41 datasets found
  1. F

    Japanese General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Japanese General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-japanese-japan
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Japanese General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Japanese speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Japanese communication.

    Curated by FutureBeeAI, this 40 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Japanese speech models that understand and respond to authentic Japanese accents and dialects.

    Speech Data

    The dataset comprises 40 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Japanese. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 80 verified native Japanese speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Japan to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Japanese speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Japanese.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Japanese conversations.

  2. F

    German General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). German General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-german-germany
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the German General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of German speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world German communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade German speech models that understand and respond to authentic German accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of German. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native German speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Germany to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple German speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for German.
    •
    Voice Assistants: Build smart assistants capable of understanding natural German conversations.
    <span

  3. F

    Finnish General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Finnish General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-finnish-finland
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Finnish General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Finnish speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Finnish communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Finnish speech models that understand and respond to authentic Finnish accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Finnish. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Finnish speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Finland to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Finnish speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Finnish.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Finnish conversations.
    <span

  4. F

    Polish General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Polish General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-polish-poland
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Area covered
    Poland
    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Polish General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Polish speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Polish communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Polish speech models that understand and respond to authentic Polish accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Polish. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Polish speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Poland to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Polish speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Polish.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Polish conversations.
    <span

  5. F

    British English General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). British English General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-english-uk
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Area covered
    United Kingdom
    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the UK English General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of English speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world UK English communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade English speech models that understand and respond to authentic British accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of UK English. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native UK English speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of United Kingdom to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple English speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for UK English.
    •
    Voice Assistants: Build smart assistants capable of understanding natural British conversations.

  6. F

    Korean General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Korean General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-korean-southkorea
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Korean General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Korean speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Korean communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Korean speech models that understand and respond to authentic Korean accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Korean. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Korean speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of South Korea to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Korean speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Korean.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Korean conversations.
    <span

  7. F

    Dutch General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Dutch General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-dutch-netherlands
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Dutch General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Dutch speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Dutch communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Dutch speech models that understand and respond to authentic Dutch accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Dutch. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Dutch speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Netherlands to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Dutch speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Dutch.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Dutch conversations.
    <span

  8. F

    American English General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). American English General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-english-usa
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Area covered
    United States
    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the US English General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of English speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world US English communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade English speech models that understand and respond to authentic American accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of US English. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native US English speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of United States of America to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple English speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for US English.
    •
    Voice Assistants: Build smart assistants capable of understanding natural American conversations.
    <div style="margin-top:10px; margin-bottom: 10px; padding-left: 30px; display: flex; gap: 16px; align-items:

  9. F

    Ukrainian General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Ukrainian General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-ukrainian-ukraine
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Area covered
    Ukraine
    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Ukrainian General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Ukrainian speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Ukrainian communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Ukrainian speech models that understand and respond to authentic Ukrainian accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Ukrainian. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Ukrainian speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Ukraine to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Ukrainian speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Ukrainian.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Ukrainian conversations.

  10. F

    Swiss German General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Swiss German General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-swiss-german
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Area covered
    Switzerland
    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Swiss German General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of German speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Swiss German communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade German speech models that understand and respond to authentic Swiss accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Swiss German. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Swiss German speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Switzerland to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple German speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Swiss German.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Swiss conversations.

  11. F

    Turkish General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Turkish General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-turkish-turkey
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Turkish General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Turkish speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Turkish communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Turkish speech models that understand and respond to authentic Turkish accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Turkish. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Turkish speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Turkey to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Turkish speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Turkish.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Turkish conversations.
    <span

  12. F

    Russian General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Russian General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-russian-russia
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Russian General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Russian speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Russian communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Russian speech models that understand and respond to authentic Russian accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Russian. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Russian speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Russia to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Russian speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Russian.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Russian conversations.
    <span

  13. F

    Norwegian General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Norwegian General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-norwegian-norway
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Norwegian General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Norwegian speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Norwegian communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Norwegian speech models that understand and respond to authentic Norwegian accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Norwegian. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Norwegian speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Norway to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Norwegian speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Norwegian.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Norwegian conversations.

  14. F

    Malay General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Malay General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-malay
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Malay General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Malay speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Malay communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Malay speech models that understand and respond to authentic Malay accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Malay. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Malay speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Malaysia to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Malay speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Malay.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Malay conversations.
    <span

  15. F

    Argentine Spanish General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    + more versions
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Argentine Spanish General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-spanish-argentina
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Area covered
    Argentina
    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Argentinians Spanish General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Spanish speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Argentinians Spanish communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Spanish speech models that understand and respond to authentic Argentinians accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Argentinians Spanish. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Argentinians Spanish speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Argentina to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Spanish speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Argentinians Spanish.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Argentinians conversations.
    <div style="margin-top:10px; margin-bottom: 10px; padding-left:

  16. F

    Swedish General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Swedish General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-swedish-sweden
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Area covered
    Sweden
    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Swedish General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Swedish speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Swedish communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Swedish speech models that understand and respond to authentic Swedish accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Swedish. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Swedish speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Sweden to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Swedish speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Swedish.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Swedish conversations.
    <span

  17. F

    Mexican Spanish General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Mexican Spanish General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-spanish-mexico
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Area covered
    Mexico
    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Mexican Spanish General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Spanish speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Mexican Spanish communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Spanish speech models that understand and respond to authentic Mexican accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Mexican Spanish. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Mexican Spanish speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Mexico to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Spanish speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Mexican Spanish.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Mexican conversations.
    <div style="margin-top:10px; margin-bottom: 10px; padding-left: 30px; display: flex; gap: 16px;

  18. F

    Canadian French General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    + more versions
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Canadian French General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-canadian-french
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Area covered
    French, Canada
    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Canadian French General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of French speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Canadian French communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade French speech models that understand and respond to authentic Canadian accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Canadian French. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Canadian French speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Canada to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple French speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Canadian French.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Canadian conversations.
    <div style="margin-top:10px; margin-bottom: 10px; padding-left: 30px; display: flex; gap: 16px;

  19. F

    Indian English General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    + more versions
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Indian English General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-english-india
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Indian English General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of English speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Indian English communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade English speech models that understand and respond to authentic Indian accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Indian English. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Indian English speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of India to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple English speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Indian English.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Indian conversations.
    <div style="margin-top:10px; margin-bottom: 10px; padding-left: 30px; display: flex; gap: 16px; align-items:

  20. F

    Filipino General Conversation Speech Dataset for ASR

    • futurebeeai.com
    wav
    Updated Aug 1, 2022
    + more versions
    Share
    FacebookFacebook
    TwitterTwitter
    Email
    Click to copy link
    Link copied
    Close
    Cite
    FutureBee AI (2022). Filipino General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-filipino-philippines
    Explore at:
    wavAvailable download formats
    Dataset updated
    Aug 1, 2022
    Dataset provided by
    FutureBeeAI
    Authors
    FutureBee AI
    License

    https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

    Area covered
    Philippines
    Dataset funded by
    FutureBeeAI
    Description

    Introduction

    Welcome to the Filipino General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Filipino speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Filipino communication.

    Curated by FutureBeeAI, this 30 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Filipino speech models that understand and respond to authentic Filipino accents and dialects.

    Speech Data

    The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Filipino. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

    •Participant Diversity:
    •
    Speakers: 60 verified native Filipino speakers from FutureBeeAI’s contributor community.
    •
    Regions: Representing various provinces of Philippines to ensure dialectal diversity and demographic balance.
    •
    Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
    •Recording Details:
    •
    Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
    •
    Duration: Each conversation ranges from 15 to 60 minutes.
    •
    Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
    •
    Environment: Quiet, echo-free settings with no background noise.

    Topic Diversity

    The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

    •Sample Topics Include:
    •Family & Relationships
    •Food & Recipes
    •Education & Career
    •Healthcare Discussions
    •Social Issues
    •Technology & Gadgets
    •Travel & Local Culture
    •Shopping & Marketplace Experiences, and many more.

    Transcription

    Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

    •Transcription Highlights:
    •Speaker-segmented dialogues
    •Time-coded utterances
    •Non-speech elements (pauses, laughter, etc.)
    •High transcription accuracy, achieved through double QA pass, average WER < 5%

    These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

    Metadata

    The dataset comes with granular metadata for both speakers and recordings:

    •
    Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
    •
    Recording Metadata: Topic, duration, audio format, device type, and sample rate.

    Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

    Usage and Applications

    This dataset is a versatile resource for multiple Filipino speech and language AI applications:

    •
    ASR Development: Train accurate speech-to-text systems for Filipino.
    •
    Voice Assistants: Build smart assistants capable of understanding natural Filipino conversations.

Share
FacebookFacebook
TwitterTwitter
Email
Click to copy link
Link copied
Close
Cite
FutureBee AI (2022). Japanese General Conversation Speech Dataset for ASR [Dataset]. https://www.futurebeeai.com/dataset/speech-dataset/general-conversation-japanese-japan

Japanese General Conversation Speech Dataset for ASR

Japanese General Conversation Speech Corpus

Explore at:
wavAvailable download formats
Dataset updated
Aug 1, 2022
Dataset provided by
FutureBeeAI
Authors
FutureBee AI
License

https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement

Dataset funded by
FutureBeeAI
Description

Introduction

Welcome to the Japanese General Conversation Speech Dataset — a rich, linguistically diverse corpus purpose-built to accelerate the development of Japanese speech technologies. This dataset is designed to train and fine-tune ASR systems, spoken language understanding models, and generative voice AI tailored to real-world Japanese communication.

Curated by FutureBeeAI, this 40 hours dataset offers unscripted, spontaneous two-speaker conversations across a wide array of real-life topics. It enables researchers, AI developers, and voice-first product teams to build robust, production-grade Japanese speech models that understand and respond to authentic Japanese accents and dialects.

Speech Data

The dataset comprises 40 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of Japanese. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.

•Participant Diversity:
•
Speakers: 80 verified native Japanese speakers from FutureBeeAI’s contributor community.
•
Regions: Representing various provinces of Japan to ensure dialectal diversity and demographic balance.
•
Demographics: A balanced gender ratio (60% male, 40% female) with participant ages ranging from 18 to 70 years.
•Recording Details:
•
Conversation Style: Unscripted, spontaneous peer-to-peer dialogues.
•
Duration: Each conversation ranges from 15 to 60 minutes.
•
Audio Format: Stereo WAV files, 16-bit depth, recorded at 16kHz sample rate.
•
Environment: Quiet, echo-free settings with no background noise.

Topic Diversity

The dataset spans a wide variety of everyday and domain-relevant themes. This topic diversity ensures the resulting models are adaptable to broad speech contexts.

•Sample Topics Include:
•Family & Relationships
•Food & Recipes
•Education & Career
•Healthcare Discussions
•Social Issues
•Technology & Gadgets
•Travel & Local Culture
•Shopping & Marketplace Experiences, and many more.

Transcription

Each audio file is paired with a human-verified, verbatim transcription available in JSON format.

•Transcription Highlights:
•Speaker-segmented dialogues
•Time-coded utterances
•Non-speech elements (pauses, laughter, etc.)
•High transcription accuracy, achieved through double QA pass, average WER < 5%

These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.

Metadata

The dataset comes with granular metadata for both speakers and recordings:

•
Speaker Metadata: Age, gender, accent, dialect, state/province, and participant ID.
•
Recording Metadata: Topic, duration, audio format, device type, and sample rate.

Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.

Usage and Applications

This dataset is a versatile resource for multiple Japanese speech and language AI applications:

•
ASR Development: Train accurate speech-to-text systems for Japanese.
•
Voice Assistants: Build smart assistants capable of understanding natural Japanese conversations.

Search
Clear search
Close search
Google apps
Main menu