Dataset

This dataset contains positive , negative and notr sentences from several data sources given in the references. In the most sentiment models , there are only two labels; positive and negative. However , user input can be totally notr sentence. For such cases there were no data I could find. Therefore I created this dataset with 3 class. Positive and negative sentences are listed below. Notr examples are extraced from turkish wiki dump. In addition, added some random text… See the full description on the dataset page: https://huggingface.co/datasets/winvoker/turkish-sentiment-analysis-dataset.

Introducing the Turkish Dialog Dataset

The Turkish Dialog Dataset is a new resource for researchers and developers working on natural language processing (NLP) and machine learning (ML) projects. This dataset contains a large collection of conversational data in Turkish, providing a valuable resource for training and testing NLP and ML models.

The dataset includes conversations from a variety of sources, including translated Cornell Movie Dialog dataset, Ubuntu Dialog dataset, speacial datasets. The data has been carefully curated and annotated to ensure high quality and accuracy.

One of the key features of the Turkish Dialog Dataset is its focus on real-world conversational data. This makes it an ideal resource for developing NLP and ML models that can understand and generate natural-sounding Turkish text.

This dataset can be used to train more sophisticated models that can understand the context of a conversation.

Overall, the Turkish Dialog Dataset is an exciting new resource for anyone working on NLP or ML projects in Turkish. Its large size and high quality make it an invaluable tool for developing advanced models that can understand and generate natural-sounding Turkish text.

Dataset Summary

This dataset is enhanced version of existing offensive language studies. Existing studies are highly imbalanced, and solving this problem is too costly. To solve this, we proposed contextual data mining method for dataset augmentation. Our method is basically prevent us from retrieving random tweets and label individually. We can directly access almost exact hate related tweets and label them directly without any further human interaction in order to solve imbalanced label problem.

In addition, existing studies (can be found at Reference section) are merged to create even more comprehensive and robust dataset for Turkish offensive language detection task.

The file train.csv contains 42,398, test.csv contains 8,851, valid.csv contains 1,756 annotated tweets.

Task Code

https://github.com/toygarr/lingda

Dataset Structure

A binary dataset with with (0) Not Offensive and (1) Offensive tweets.

Task and Labels

Offensive language identification: (0) Not Offensive - Tweet does not contain offense or profanity. (1) Offensive - Tweet contains offensive language or a targeted (veiled or direct) offense

Data Splits

Citation Information

T. Tanyel, B. Alkurdi and S. Ayvaz, "Linguistic-based Data Augmentation Approach for Offensive Language Detection," 2022 7th International Conference on Computer Science and Engineering (UBMK), 2022, pp. 1-6, doi: 10.1109/UBMK55850.2022.9919562

References

We merged open-source offensive language dataset studies in Turkish to increase contextuality with existing data even more, before our method is applied. - https://huggingface.co/datasets/offenseval2020_tr - https://github.com/imayda/turkish-hate-speech-dataset-2 - https://www.kaggle.com/datasets/kbulutozler/5k-turkish-tweets-with-incivil-content

NLI-TR (Turkish NLI Research)

Unleash Your NLI Research in Turkish Language!

By Huggingface Hub [source]

About this dataset

NLI-TR is a revolutionary set of two datasets that provide an unparalleled opportunity for the natural language processing and machine learning community to conduct inference research in the Turkish Language. The datasets - SNLI-TR and MNLI-TR - contain carefully curated natural language inference data that have been translated into Turkish. With NLI-TR, researchers can explore the exciting prospects of developing automated models tailored to make inferences on texts produced in this vibrant language. Moreover, they can also investigate how models trained on data from one language fare when applied in another, a valuable insight into cross-lingual generalization capabilities. NLI-TR offers both seasoned and budding researchers an unprecedented platform to further our understanding of natural language inferencing capability

More Datasets

For more datasets, click here.

Featured Notebooks

• 🚨 Your notebook can be here! 🚨!

How to use the dataset

How To Use The NLI-TR Dataset to Unlock Turkish NLI Research

Welcome to the exciting world of natural language inference (NLI) research! If you’re looking for a great dataset to use for your research in this field, the NLI-TR dataset is a perfect starting point. This guide will provide an overview of how you can use the data from this dataset to uncover new insights about NLI tasks in Turkish.

The NLI-TR dataset contains two large scale datasets intended for natural language inference tasks – SNLI-TR and MNLI- TR. Both datasets offer researchers an opportunity to explore Natural Language Inference (NLI) research in the Turkish language, with examples ranging from sentence paraphrasing task and classification tasks to question answering scenarios using various NLP techniques.

Using the Data:

The data provided in this dataset includes both training and validation sets, making it easy for researchers who are just getting started with their projects. The SNLI_tr_train.csv file is used as input for training your models, while slni_tr_validation can be used as input for testing or validating model accuracy on unseen data. Additionally, multinli_tr_validation_{matched / mismatched}.csv files offer additional validation on how well your trained models perform on more complex scenarios such as sentence paraphrasing or question answering tasks using various NLP techniques.

Each record includes four columns – premise ,hypothesis ,label , (and domain). The premise column specifies what information is provided before asking a question or making an inference; think of it as context clues that explain why one statement implies another statement more directly than others might do without them . The hypothesis column provides what lies at the heart of inference --the conclusion reached after introducing facts given before it . Last but not least we have label column which denotes whether two sentences entail each other (ENTAILMENT), contradict each other(CONTRADICTION) or are unrelated(NEUTRAL). A domain label has also been assigned by some authors when necessary; this mostly applies when inferring between sentences across different semantic domains such as weather vs sports vs finance etc .

Research Ideas

• Developing an NLI-based Turkish language question answering system.
• Training a sentiment analysis algorithm to identify sentiment in text written in Turkish.
• Building a Machine Learning Chatbot that uses NLI to understand conversational context and respond accordingly for users intending to converse in the Turkish language

Acknowledgements

If you use this dataset in your research, please credit the original authors. Data Source

License

License: CC0 1.0 Universal (CC0 1.0) - Public Domain Dedication No Copyright - You can copy, modify, distribute and perform the work, even for commercial purposes, all without asking permission. See Other Information.

Columns

File: snli_tr_train.csv | Column name | Description | |:---------------|:------------------------------------------------------------------------------------------------------------------------------------------...

Dataset Summary

This dataset is enhanced version of existing offensive language studies. Existing studies are highly imbalanced, and solving this problem is too costly. To solve this, we proposed contextual data mining method for dataset augmentation. Our method is basically prevent us from retrieving random tweets and label individually. We can directly access almost exact hate related tweets and label them directly without any further human interaction in order to solve imbalanced… See the full description on the dataset page: https://huggingface.co/datasets/Toygar/turkish-offensive-language-detection.

TURKISH SONG LYRICS FROM GENIUS DATASET

DATASET DESCRIPTION

This dataset contains a comprehensive collection of 44,692 Turkish song lyrics, extracted from the larger "Genius Song Lyrics with Language Information" dataset available on Kaggle. The original 9.07 GB dataset was filtered to include only songs identified with the language code 'tr' (Turkish), making it a clean and focused resource for Turkish Natural Language Processing (NLP) tasks.

[TR] Bu veri seti, Kaggle'da bulunan "Genius Song Lyrics with Language Information" adlı büyük veri setinden ayıklanmış 44,692 adet Türkçe şarkı sözü içermektedir. Orijinal 9.07 GB'lık veri seti, dil kodu 'tr' (Türkçe) olarak tanımlanmış şarkıları içerecek şekilde filtrelenmiştir. Bu, Türkçe Doğal Dil İşleme (DDİ) görevleri için temiz ve odaklanmış bir kaynak oluşturmaktadır.

HOW TO USE

You can easily load this dataset using the Hugging Face datasets library.

[TR] Bu veri setini Hugging Face datasets kütüphanesini kullanarak kolayca yükleyebilirsiniz.

Örnek Python Kodu: ```python from datasets import load_dataset

Load the dataset from the Hugging Face Hub

Veri setini Hugging Face Hub'dan yükleyin

dataset = load_dataset("mustafakemal0146/Genius-Turkish-Dataset")

Example: Access the lyrics of the first song in the training split

Örnek: Eğitim setindeki ilk şarkının sözlerine erişim

print(dataset['train'][0]['lyrics']) ```

DATASET STRUCTURE

The dataset consists of a single CSV file, loaded as the train split, with the following columns:

[TR] Veri seti, train bölünmüşü olarak yüklenen ve aşağıdaki sütunları içeren tek bir CSV dosyasından oluşur:

• title (string): The title of the song / Şarkının başlığı.
• tag (string): The genre tag associated with the song (e.g., 'rap', 'pop') / Şarkıyla ilişkilendirilen tür etiketi.
• artist (string): The name of the primary artist / Ana sanatçının adı.
• year (int): The release year of the song / Şarkının çıkış yılı.
• views (int): The number of views on Genius.com / Genius.com'daki görüntülenme sayısı.
• features (string): A string representation of featuring artists / Düet yapılan sanatçıların metin formatı.
• lyrics (string): The full lyrics of the song / Şarkının tam sözleri.
• id (int): A unique identifier from the original dataset / Orijinal veri setinden gelen benzersiz ID.
• language_cld3 (string): Language code detected by CLD3 model (all 'tr') / CLD3 ile tespit edilen dil kodu (tümü 'tr').
• language_ft (string): Language code detected by FastText model (all 'tr') / FastText ile tespit edilen dil (tümü 'tr').
• language (string): Final aggregated language code (all 'tr') / Nihai dil kodu (tümü 'tr').

DATA SOURCE AND CURATION

This dataset is a curated subset of the "Genius Song Lyrics with Language Information" (Link: https://www.kaggle.com/datasets/pavanelisetty/genius-song-lyrics-with-language-information) dataset on Kaggle, originally collected from Genius.com. The filtering process involved reading the main song_lyrics.csv file and selecting all rows where the language column was equal to 'tr'.

[TR] Bu veri seti, orijinal olarak Genius.com'dan toplanmış olan Kaggle'daki "Genius Song Lyrics with Language Information" veri setinin düzenlenmiş bir alt kümesidir. Filtreleme işlemi, ana song_lyrics.csv dosyasını okuyarak language sütununun 'tr' olduğu tüm satırların seçilmesiyle yapılmıştır.

LICENSE

The original dataset on Kaggle does not specify a license. This curated version is shared under the "Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0)" (Link: https://creativecommons.org/licenses/by-nc-sa/4.0/) license, assuming it will be used for non-commercial research and educational purposes. Please refer to the original data source for any commercial use inquiries.

Created by MustafaKemal0146 (Hugging Face Profile: https://huggingface.co/MustafaKemal0146)

Turkish Number Plates

## Overview

Turkish Number Plates is a dataset for object detection tasks - it contains Plate annotations for 2,246 images.

## Getting Started

You can download this dataset for use within your own projects, or fork it into a workspace on Roboflow to create your own model.

  ## License

  This dataset is available under the [CC BY 4.0 license](https://creativecommons.org/licenses/CC BY 4.0).

This dataset contains 1620 hours of Turkish speech collected from real-world conversational and monologue scenarios. Each audio recording includes accurate transcripts, speaker ID, gender, and additional metadata. The dataset was collected from diverse Turkish speakers with different geographic and demographic backgrounds, enhancing model performance in real and complex tasks. Quality tested by various AI companies. We strictly adhere to data protection regulations and privacy standards, ensuring the maintenance of user privacy and legal rights throughout the data collection, storage, and usage processes, our datasets are all GDPR, CCPA, PIPL complied.

Turkish Speech Corpus (TSC)

This repository presents an open-source Turkish Speech Corpus, introduced in "Multilingual Speech Recognition for Turkic Languages". The corpus contains 218.2 hours of transcribed speech with 186,171 utterances and is the largest publicly available Turkish dataset of its kind at that time. Paper: Multilingual Speech Recognition for Turkic Languages.
GitHub Repository: https://github.com/IS2AI/TurkicASR

  Citation

@Article{info14020074… See the full description on the dataset page: https://huggingface.co/datasets/issai/Turkish_Speech_Corpus.

The Turkish TTS Monologue Speech Dataset is a professionally curated resource built to train realistic, expressive, and production-grade text-to-speech (TTS) systems. It contains studio-recorded long-form speech by trained native Turkish voice artists, each contributing 1 to 2 hours of clean, uninterrupted monologue audio.

Unlike typical prompt-based datasets with short, isolated phrases, this collection features long-form, topic-driven monologues that mirror natural human narration. It includes content types that are directly useful for real-world applications, like audiobook-style storytelling, educational lectures, health advisories, product explainers, digital how-tos, formal announcements, and more.

All recordings are captured in professional studios using high-end equipment and under the guidance of experienced voice directors.

Recording & Audio Quality

Audio Format: WAV, 48 kHz, available in 16-bit, 24-bit, and 32-bit depth

SNR: Minimum 30 dB

Channel: Mono

Recording Duration: 20-30 minutes

Recording Environment: Studio-controlled, acoustically treated rooms

Per Speaker Volume: 1–2 hours of speech per artist

Quality Control: Each file is reviewed and cleaned for common acoustic issues, including: reverberation, lip smacks, mouth clicks, thumping, hissing, plosives, sibilance, background noise, static interference, clipping, and other artifacts.

Only clean, production-grade audio makes it into the final dataset.

Voice Artist Selection

All voice artists are native Turkish speakers with professional training or prior experience in narration. We ensure a diverse pool in terms of age, gender, and region to bring a balanced and rich vocal dataset.

Artist Profile:

- Gender: Male and Female

- Age Range: 20–60 years

- Regions: Native Turkish-speaking states from Turkey

Selection Process: All artists are screened, onboarded, and sample-approved using FutureBeeAI’s proprietary Yugo platform.

Script Quality & Coverage

Scripts are not generic or repetitive. Scripts are professionally authored by domain experts to reflect real-world use cases. They avoid redundancy and include modern vocabulary, emotional range, and phonetically rich sentence structures.

Word Count per Script: 3,000–5,000 words per 30-minute session

Content Types:

- Storytelling

- Script and book reading

- Informational explainers

- Government service instructions

- E-commerce tutorials

- Motivational content

- Health & wellness guides

- Education & career advice

Linguistic Design: Balanced punctuation, emotional range, modern syntax, and vocabulary diversity

Transcripts & Alignment

While the script is used during the recording, we also provide post-recording updates to ensure the transcript reflects the final spoken audio. Minor edits are made to adjust for skipped or rephrased words.

Segmentation: Time-stamped at the sentence level, aligned to actual spoken delivery

Format: Available in plain text and JSON

Post-processing:

- Corrected for disfluencies

- Matched to actual spoken audio

- Quality checked by a native language expert

Metadata

Every recording and speaker is tagged with detailed metadata. This metadata enables filtering, speaker-level analysis, and targeted fine-tuning during model training.

Speaker Metadata:

- Gender

- Age group

- Regional background

Recording Metadata:

- File duration

- Bit depth

- Sample rate

- Acoustic environment

- QA status

Content Metadata:

- Domain (e.g., health, education, product guide)

- Topic

- Speaking style (neutral, formal, emotional)

Quality Assurance Process

Each recording goes through a multi-stage QA workflow:

Pre-Recording: Voice artists are trained on script delivery, tone, and clarity

Supervised Recording: Emotional or expressive scripts are recorded with guidance from voice directors

Linguistic Review: Native Turkish experts review the spoken content for accuracy and delivery

Acoustic Review: Sound engineers evaluate each file to detect and clean technical artifacts

Final Approval: Only files that pass all checks and include complete metadata are added to the final dataset.

License & Usage Rights

The Turkish TTS Monologue Speech Dataset is developed by FutureBeeAI and is available exclusively under a commercial license, designed to support a wide range of real-world applications while ensuring ethical and legal compliance.

Introduction

This Turkish Call Center Speech Dataset for the Delivery and Logistics industry is purpose-built to accelerate the development of speech recognition, spoken language understanding, and conversational AI systems tailored for Turkish-speaking customers. With over 30 hours of real-world, unscripted call center audio, this dataset captures authentic delivery-related conversations essential for training high-performance ASR models.

Curated by FutureBeeAI, this dataset empowers AI teams, logistics tech providers, and NLP researchers to build accurate, production-ready models for customer support automation in delivery and logistics.

Speech Data

The dataset contains 30 hours of dual-channel call center recordings between native Turkish speakers. Captured across various delivery and logistics service scenarios, these conversations cover everything from order tracking to missed delivery resolutions offering a rich, real-world training base for AI models.

Participant Diversity:

- Speakers: 60 native Turkish speakers from our verified contributor pool.

- Regions: Multiple provinces of Turkey for accent and dialect diversity.

- Participant Profile: Balanced gender distribution (60% male, 40% female) with ages ranging from 18 to 70.

Recording Details:

- Conversation Nature: Naturally flowing, unscripted customer-agent dialogues.

- Call Duration: 5 to 15 minutes on average.

- Audio Format: Stereo WAV, 16-bit depth, recorded at 8kHz and 16kHz.

- Recording Environment: Captured in clean, noise-free, echo-free conditions.

Topic Diversity

This speech corpus includes both inbound and outbound delivery-related conversations, covering varied outcomes (positive, negative, neutral) to train adaptable voice models.

Inbound Calls:

- Order Tracking

- Delivery Complaints

- Undeliverable Addresses

- Return Process Enquiries

- Delivery Method Selection

- Order Modifications, and more

Outbound Calls:

- Delivery Confirmations

- Subscription Offer Calls

- Incorrect Address Follow-ups

- Missed Delivery Notifications

- Delivery Feedback Surveys

- Out-of-Stock Alerts, and others

This comprehensive coverage reflects real-world logistics workflows, helping voice AI systems interpret context and intent with precision.

Transcription

All recordings come with high-quality, human-generated verbatim transcriptions in JSON format.

Transcription Includes:

- Speaker-Segmented Dialogues

- Time-coded Segments

- Non-speech Tags (e.g., pauses, noise)

- High transcription accuracy with word error rate under 5% via dual-layer quality checks.

These transcriptions support fast, reliable model development for Turkish voice AI applications in the delivery sector.

Metadata

Detailed metadata is included for each participant and conversation:

Participant Metadata: ID, age, gender, region, accent, dialect.

Conversation Metadata: Topic, call type, sentiment, sample rate, and technical attributes.

This metadata aids in training specialized models, filtering demographics, and running advanced analytics.

License

This Delivery and Logistics domain dataset is commercially licensed and ready for use in ASR, NLP, and voice automation projects in Turkish.

This dataset contains positive, negative, and neutral sentences from several data sources in the references. In most sentiment models, there are only two labels; positive and negative. However, user input can be totally neutral sentence. For such cases, there were no data I could find. Therefore I created this dataset with 3 classes. Positive and negative sentences are listed below. neutral examples are extracted from Turkish wiki dump. In addition, I added some random text inputs like "Lorem ipsum dolor sit amet".

There are 492.782 labeled sentences. %10 of them were used for testing.

This dataset is also available at Hugging Face. You can directly use it. https://huggingface.co/datasets/winvoker/turkish-sentiment-analysis-dataset

References - https://www.kaggle.com/burhanbilenn/duygu-analizi-icin-urun-yorumlari - https://github.com/fthbrmnby/turkish-text-data - https://www.kaggle.com/mustfkeskin/turkish-wikipedia-dump - https://github.com/ezgisubasi/turkish-tweets-sentiment-analysis - http://humirapps.cs.hacettepe.edu.tr/

Stanford alpaca turkish: Stanford Alpaca

Turkish Book Data Set

This dataset is a comprehensive compilation of Turkish books obtained through web scraping from the internet. Each record in the dataset contains essential information such as the book's title, author, publisher, publication year, page count, category, description, and image URL.

This rich dataset can be utilized in various applications, particularly in the analysis through methods such as classification, content-based recommendation algorithms, and natural language processing (NLP). For researchers, students, and data scientists, this dataset serves as a valuable resource for exploring Turkish literature, generating book recommendations, or developing machine learning models.

Enhance your Conversational AI model with our Off-the-Shelf Turkish Language Dataset (Turkish Language Speech Datasets). Shaip high-quality audio datasets are a quick

Evaluation of Multi-label Target Classification performance on the Turkish Dataset, measured using Precision, Recall, F1-score, and Support.

Dataset Card for "turkishneuralvoice"

  Dataset Overview

Dataset Name: Turkish Neural Voice Description: This dataset contains Turkish audio samples generated using Microsoft Text to Speech services. The dataset includes audio files and their corresponding transcriptions.

  Dataset Structure

Configs:

default

Data Files:

Split: train Path: data/train-*

Dataset Info:

Features: audio: Audio file transcription: Corresponding text transcription

Splits: train… See the full description on the dataset page: https://huggingface.co/datasets/erenfazlioglu/turkishvoicedataset.

Turkish MMLU is a Turkish-language multiple-choice question dataset containing 293,468 samples spanning 800 topics organized across 67 sections. Created by M. Ali Bayram and published in 2024, it is designed to support fine-tuning, evaluation, and research involving large language models in the Turkish language.

The dataset is an original resource intended to serve as a Turkish equivalent of the Massive Multitask Language Understanding (MMLU) benchmark, covering educational and general knowledge domains relevant to Turkish-speaking contexts. It is available under a CC BY-NC 4.0 license, meaning it is restricted to non-commercial use, and is accessible via Hugging Face with a registered DOI through Zenodo.

About this dataset\(\)\(\)\(\)\(**Context**\)\(\)\(\)\(\)This list ranks the 37 cities in the Western United States by Turkish population, as estimated by the United States Census Bureau. It also highlights population changes in each city over the past five years.\(\)\(\)\(\)\(\)\(\)$###### Content\(\)\(\)\(\)\(**When available, the data consists of estimates from the U.S. Census Bureau American Community Survey (ACS) 5-Year Estimates, including:**\)\(\)\(\)\(\)- 2020-2024 American Community Survey 5-Year Estimates\(\)\($- 2014-2018 American Community Survey 5-Year Estimates\)\(\)$- 2009-2013 American Community Survey 5-Year Estimates\(\)\(\)\(\)\(**Variables / Data Columns**\)\(\)\(\)\(\)- Rank by Turkish Population: This column displays the rank of city in the Western United States by their Turkish population, using the most recent ACS data available.\(\)\($- *City:* The city for which the rank is shown in the previous column.\)\(\)$- Turkish Population: The Turkish population of the city is shown in this column.\(\)\($- *% of Total City Population:* This shows what percentage of the total city population identifies as Turkish. Please note that the sum of all percentages may not equal one due to rounding of values.\)\(\)$- % of Total Western United States Turkish Population: This tells us how much of the entire Western United States Turkish population lives in that city. Please note that the sum of all percentages may not equal one due to rounding of values.\(\)\($- *5 Year Rank Trend:* This column displays the rank trend across the last 5 years.\)\(\)\(\)\(\)Good to know\(\)\(\)\(\)\(*Margin of Error*\)\(\)\(\)\(\)Data in the dataset are based on the estimates and are subject to sampling variability and thus a margin of error. Neilsberg Research recommends using caution when presening these estimates in your research.\(\)\(\)\(\)\(*Custom data*\)\(\)\(\)\(\)If you do need custom data for any of your research project, report or presentation, you can contact our research staff at research@neilsberg.com for a feasibility of a custom tabulation on a fee-for-service basis.\(\)\(\)\(\)\(\)\(\)$###### Inspiration\(\)\(\)\(\)$$Neilsberg Research Team curates, analyze and publishes demographics and economic data from a variety of public and proprietary sources, each of which often includes multiple surveys and programs. The large majority of Neilsberg Research aggregated datasets and insights is made available for free download at https://www.neilsberg.com/research/.

Turkish-SFT-Dataset-v1.01

Repo: AlicanKiraz0/Turkish-SFT-Dataset-v1.0Sürüm: v1.01Lisans: MITBiçim: jsonl (kolonlar: system, user, assistant)Boyut: ~5500 satır ve satır başına 3.000–4.500 token/satır (≈ 20M+ token)Dil: Türkçe (tr)Görevler: talim...