Mexico is the country with the largest number of native Spanish speakers in the world. As of 2024, 132.5 million people in Mexico spoke Spanish with a native command of the language. Colombia was the nation with the second-highest number of native Spanish speakers, at around 52.7 million. Spain came in third, with 48 million, and Argentina fourth, with 46 million. Spanish, a world language As of 2023, Spanish ranked as the fourth most spoken language in the world, only behind English, Chinese, and Hindi, with over half a billion speakers. Spanish is the official language of over 20 countries, the majority on the American continent, nonetheless, it's also one of the official languages of Equatorial Guinea in Africa. Other countries have a strong influence, like the United States, Morocco, or Brazil, countries included in the list of non-Hispanic countries with the highest number of Spanish speakers. The second most spoken language in the U.S. In the most recent data, Spanish ranked as the language, other than English, with the highest number of speakers, with 12 times more speakers as the second place. Which comes to no surprise following the long history of migrations from Latin American countries to the Northern country. Moreover, only during the fiscal year 2022. 5 out of the top 10 countries of origin of naturalized people in the U.S. came from Spanish-speaking countries.
The United States is the non-hispanic country with the largest number of native Spanish speakers in the world, with approximately 41.89 million people with a native command of the language in 2024. However, the European Union had the largest group of non-native speakers with limited proficiency of Spanish, at around 28 million people. Furthermore, Mexico is the country with the largest number of native Spanish speakers in the world as of 2024.
In 2023, around 43.37 million people in the United States spoke Spanish at home. In comparison, approximately 998,179 people were speaking Russian at home during the same year. The distribution of the U.S. population by ethnicity can be accessed here. A ranking of the most spoken languages across the world can be accessed here.
In 2025, there were around 1.53 billion people worldwide who spoke English either natively or as a second language, slightly more than the 1.18 billion Mandarin Chinese speakers at the time of survey. Hindi and Spanish accounted for the third and fourth most widespread languages that year. Languages in the United States The United States does not have an official language, but the country uses English, specifically American English, for legislation, regulation, and other official pronouncements. The United States is a land of immigration, and the languages spoken in the United States vary as a result of the multicultural population. The second most common language spoken in the United States is Spanish or Spanish Creole, which over than 43 million people spoke at home in 2023. There were also 3.5 million Chinese speakers (including both Mandarin and Cantonese),1.8 million Tagalog speakers, and 1.57 million Vietnamese speakers counted in the United States that year. Different languages at home The percentage of people in the United States speaking a language other than English at home varies from state to state. The state with the highest percentage of population speaking a language other than English is California. About 45 percent of its population was speaking a language other than English at home in 2023.
In 2023, California had the highest Hispanic population in the United States, with over 15.76 million people claiming Hispanic heritage. Texas, Florida, New York, and Illinois rounded out the top five states for Hispanic residents in that year. History of Hispanic people Hispanic people are those whose heritage stems from a former Spanish colony. The Spanish Empire colonized most of Central and Latin America in the 15th century, which began when Christopher Columbus arrived in the Americas in 1492. The Spanish Empire expanded its territory throughout Central America and South America, but the colonization of the United States did not include the Northeastern part of the United States. Despite the number of Hispanic people living in the United States having increased, the median income of Hispanic households has fluctuated slightly since 1990. Hispanic population in the United States Hispanic people are the second-largest ethnic group in the United States, making Spanish the second most common language spoken in the country. In 2021, about one-fifth of Hispanic households in the United States made between 50,000 to 74,999 U.S. dollars. The unemployment rate of Hispanic Americans has fluctuated significantly since 1990, but has been on the decline since 2010, with the exception of 2020 and 2021, due to the impact of the coronavirus (COVID-19) pandemic.
This data set uses the 2009-2013 American Community Survey to tabulate the number of speakers of languages spoken at home and the number of speakers of each language who speak English less than very well. These tabulations are available for the following geographies: nation; each of the 50 states, plus Washington, D.C. and Puerto Rico; counties with 100,000 or more total population and 25,000 or more speakers of languages other than English and Spanish; core-based statistical areas (metropolitan statistical areas and micropolitan statistical areas) with 100,000 or more total population and 25,000 or more speakers of languages other than English and Spanish.
The United States is the country with the largest number of Spanish language students, at approximately 8.59 million people in 2024. The second country is Brazil, with around 4.05 million students of the Spanish language. Moreover, the United States is also the non-hispanic country with the largest number of native Spanish speakers in the world.
Many residents of New York City speak more than one language; a number of them speak and understand non-English languages more fluently than English. This dataset, derived from the Census Bureau's American Community Survey (ACS), includes information on over 1.7 million limited English proficient (LEP) residents and a subset of that population called limited English proficient citizens of voting age (CVALEP) at the Community District level. There are 59 community districts throughout NYC, with each district being represented by a Community Board.
CC0 1.0 Universal Public Domain Dedicationhttps://creativecommons.org/publicdomain/zero/1.0/
License information was derived automatically
Although the American Community Survey (ACS) produces population, demographic and housing unit estimates, it is the Census Bureau's Population Estimates Program that produces and disseminates the official estimates of the population for the nation, states, counties, cities, and towns and estimates of housing units for states and counties..Supporting documentation on code lists, subject definitions, data accuracy, and statistical testing can be found on the American Community Survey website in the Technical Documentation section.Sample size and data quality measures (including coverage rates, allocation rates, and response rates) can be found on the American Community Survey website in the Methodology section..Source: U.S. Census Bureau, 2019 American Community Survey 1-Year Estimates.Data are based on a sample and are subject to sampling variability. The degree of uncertainty for an estimate arising from sampling variability is represented through the use of a margin of error. The value shown here is the 90 percent margin of error. The margin of error can be interpreted roughly as providing a 90 percent probability that the interval defined by the estimate minus the margin of error and the estimate plus the margin of error (the lower and upper confidence bounds) contains the true value. In addition to sampling variability, the ACS estimates are subject to nonsampling error (for a discussion of nonsampling variability, see ACS Technical Documentation). The effect of nonsampling error is not represented in these tables..In 2016, changes were made to the languages and language categories presented in tables B16001, C16001, and B16002. For more information, see: 2016 Language Data User note..The 2019 American Community Survey (ACS) data generally reflect the September 2018 Office of Management and Budget (OMB) delineations of metropolitan and micropolitan statistical areas. In certain instances the names, codes, and boundaries of the principal cities shown in ACS tables may differ from the OMB delineations due to differences in the effective dates of the geographic entities..Estimates of urban and rural populations, housing units, and characteristics reflect boundaries of urban areas defined based on Census 2010 data. As a result, data for urban and rural areas from the ACS do not necessarily reflect the results of ongoing urbanization..Explanation of Symbols:An "**" entry in the margin of error column indicates that either no sample observations or too few sample observations were available to compute a standard error and thus the margin of error. A statistical test is not appropriate.An "-" entry in the estimate column indicates that either no sample observations or too few sample observations were available to compute an estimate, or a ratio of medians cannot be calculated because one or both of the median estimates falls in the lowest interval or upper interval of an open-ended distribution, or the margin of error associated with a median was larger than the median itself.An "-" following a median estimate means the median falls in the lowest interval of an open-ended distribution.An "+" following a median estimate means the median falls in the upper interval of an open-ended distribution.An "***" entry in the margin of error column indicates that the median falls in the lowest interval or upper interval of an open-ended distribution. A statistical test is not appropriate.An "*****" entry in the margin of error column indicates that the estimate is controlled. A statistical test for sampling variability is not appropriate. An "N" entry in the estimate and margin of error columns indicates that data for this geographic area cannot be displayed because the number of sample cases is too small.An "(X)" means that the estimate is not applicable or not available.
https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement
This US Spanish Call Center Speech Dataset for the Telecom industry is purpose-built to accelerate the development of speech recognition, spoken language understanding, and conversational AI systems tailored for Spanish-speaking telecom customers. Featuring over 30 hours of real-world, unscripted audio, it delivers authentic customer-agent interactions across key telecom support scenarios to help train robust ASR models.
Curated by FutureBeeAI, this dataset empowers voice AI engineers, telecom automation teams, and NLP researchers to build high-accuracy, production-ready models for telecom-specific use cases.
The dataset contains 30 hours of dual-channel call center recordings between native US Spanish speakers. Captured in realistic customer support settings, these conversations span a wide range of telecom topics from network complaints to billing issues, offering a strong foundation for training and evaluating telecom voice AI solutions.
This speech corpus includes both inbound and outbound calls with varied conversational outcomes like positive, negative, and neutral ensuring broad scenario coverage for telecom AI development.
This variety helps train telecom-specific models to manage real-world customer interactions and understand context-specific voice patterns.
All audio files are accompanied by manually curated, time-coded verbatim transcriptions in JSON format.
These transcriptions are production-ready, allowing for faster development of ASR and conversational AI systems in the Telecom domain.
Rich metadata is available for each participant and conversation:
https://catalogue.elra.info/static/from_media/metashare/licences/ELRA_END_USER.pdfhttps://catalogue.elra.info/static/from_media/metashare/licences/ELRA_END_USER.pdf
https://catalogue.elra.info/static/from_media/metashare/licences/ELRA_VAR.pdfhttps://catalogue.elra.info/static/from_media/metashare/licences/ELRA_VAR.pdf
The US Spanish Speecon database is divided into 2 sets: 1) The first set comprises the recordings of 550 adult Spanish speakers in the US (255 males, 295 females), recorded over 4 microphone channels in 4 recording environments (office, entertainment, car, public place), and consisting of about 208 hours of audio data. 2) The second set comprises the recordings of 50 child Spanish speakers in the US (28 boys, 22 girls), recorded over 4 microphone channels in 1 recording environment (children room), and consisting of about 14.7 hours of audio data. This database is partitioned into 22 DVDs (first set) and 3 DVDs (second set).The speech databases made within the Speecon project were validated by SPEX, the Netherlands, to assess their compliance with the Speecon format and content specifications.Each of the four speech channels is recorded at 16 kHz, 16 bit, uncompressed unsigned integers in Intel format (lo-hi byte order). To each signal file corresponds an ASCII SAM label file which contains the relevant descriptive information.Each speaker uttered the following items:Calibration data: 6 noise recordings The “silence word” recordingFree spontaneous items (adults only):2 minutes (session time) of free spontaneous, rich context items (story telling) (an open number of spontaneous topics out of a set of 30 topics)17 Elicited spontaneous items (adults only):3 dates, 2 times, 3 proper names, 2 city names, 1 letter sequence, 2 answers to questions, 3 telephone numbers, 1 language Read speech:30 phonetically rich sentences uttered by adults and 60 uttered by children5 phonetically rich words (adults only)4 isolated digits1 isolated digit sequence4 connected digit sequences1 telephone number3 natural numbers1 money amount2 time phrases (T1 : analogue, T2 : digital)3 dates (D1 : analogue, D2 : relative and general date, D3 : digital)3 letter sequences1 proper name2 city or street names2 questions2 special keyboard characters 1 Web address1 email address208 application specific words and phrases per session (adults)74 toy commands, 14 phone commands and 34 general commands (children)The following age distribution has been obtained: Adults: 223 speakers are between 15 and 30, 191 speakers are between 31 and 45, and 136 speakers are over 46.Children: 15 speakers are between 8 and 10, 35 speakers are between 11 and 14.A pronunciation lexicon with a phonemic transcription in SAMPA is also included.
This dataset contains estimates of the number of residents aged 5 years or older in Chicago who “speak English less than very well,” by the non-English language spoken at home and community area of residence, for the years 2008 – 2012. See the full dataset description for more information at: https://data.cityofchicago.org/api/views/fpup-mc9v/files/dK6ZKRQZJ7XEugvUavf5MNrGNW11AjdWw0vkpj9EGjg?download=true&filename=P:\EPI\OEPHI\MATERIALS\REFERENCES\ECONOMIC_INDICATORS\Dataset_Description_Languages_2012_FOR_PORTAL_ONLY.pdf
CC0 1.0 Universal Public Domain Dedicationhttps://creativecommons.org/publicdomain/zero/1.0/
License information was derived automatically
Although the American Community Survey (ACS) produces population, demographic and housing unit estimates, the decennial census is the official source of population totals for April 1st of each decennial year. In between censuses, the Census Bureau's Population Estimates Program produces and disseminates the official estimates of the population for the nation, states, counties, cities, and towns and estimates of housing units and the group quarters population for states and counties..Information about the American Community Survey (ACS) can be found on the ACS website. Supporting documentation including code lists, subject definitions, data accuracy, and statistical testing, and a full list of ACS tables and table shells (without estimates) can be found on the Technical Documentation section of the ACS website.Sample size and data quality measures (including coverage rates, allocation rates, and response rates) can be found on the American Community Survey website in the Methodology section..Source: U.S. Census Bureau, 2023 American Community Survey 1-Year Estimates.ACS data generally reflect the geographic boundaries of legal and statistical areas as of January 1 of the estimate year. For more information, see Geography Boundaries by Year..Data are based on a sample and are subject to sampling variability. The degree of uncertainty for an estimate arising from sampling variability is represented through the use of a margin of error. The value shown here is the 90 percent margin of error. The margin of error can be interpreted roughly as providing a 90 percent probability that the interval defined by the estimate minus the margin of error and the estimate plus the margin of error (the lower and upper confidence bounds) contains the true value. In addition to sampling variability, the ACS estimates are subject to nonsampling error (for a discussion of nonsampling variability, see ACS Technical Documentation). The effect of nonsampling error is not represented in these tables..Users must consider potential differences in geographic boundaries, questionnaire content or coding, or other methodological issues when comparing ACS data from different years. Statistically significant differences shown in ACS Comparison Profiles, or in data users' own analysis, may be the result of these differences and thus might not necessarily reflect changes to the social, economic, housing, or demographic characteristics being compared. For more information, see Comparing ACS Data..In 2016, changes were made to the languages and language categories presented in tables B16001, C16001, and B16002. For more information, see: 2016 Language Data User note..Estimates of urban and rural populations, housing units, and characteristics reflect boundaries of urban areas defined based on 2020 Census data. As a result, data for urban and rural areas from the ACS do not necessarily reflect the results of ongoing urbanization..Explanation of Symbols:- The estimate could not be computed because there were an insufficient number of sample observations. For a ratio of medians estimate, one or both of the median estimates falls in the lowest interval or highest interval of an open-ended distribution. For a 5-year median estimate, the margin of error associated with a median was larger than the median itself.N The estimate or margin of error cannot be displayed because there were an insufficient number of sample cases in the selected geographic area. (X) The estimate or margin of error is not applicable or not available.median- The median falls in the lowest interval of an open-ended distribution (for example "2,500-")median+ The median falls in the highest interval of an open-ended distribution (for example "250,000+").** The margin of error could not be computed because there were an insufficient number of sample observations.*** The margin of error could not be computed because the median falls in the lowest interval or highest interval of an open-ended distribution.***** A margin of error is not appropriate because the corresponding estimate is controlled to an independent population or housing estimate. Effectively, the corresponding estimate has no sampling error and the margin of error may be treated as zero.
This layer shows language group of language spoken at home by age. This is shown by tract, county, and state boundaries. This service is updated annually to contain the most currently released American Community Survey (ACS) 5-year data, and contains estimates and margins of error. There are also additional calculated attributes related to this topic, which can be mapped or used within analysis. This layer is symbolized to show the percentage of the population age 5+ who speak Spanish at home. To see the full list of attributes available in this service, go to the "Data" tab, and choose "Fields" at the top right. Current Vintage: 2019-2023ACS Table(s): B16007Data downloaded from: Census Bureau's API for American Community Survey Date of API call: December 12, 2024National Figures: data.census.govThe United States Census Bureau's American Community Survey (ACS):About the SurveyGeography & ACSTechnical DocumentationNews & UpdatesThis ready-to-use layer can be used within ArcGIS Pro, ArcGIS Online, its configurable apps, dashboards, Story Maps, custom apps, and mobile apps. Data can also be exported for offline workflows. For more information about ACS layers, visit the FAQ. Please cite the Census and ACS when using this data.Data Note from the Census:Data are based on a sample and are subject to sampling variability. The degree of uncertainty for an estimate arising from sampling variability is represented through the use of a margin of error. The value shown here is the 90 percent margin of error. The margin of error can be interpreted as providing a 90 percent probability that the interval defined by the estimate minus the margin of error and the estimate plus the margin of error (the lower and upper confidence bounds) contains the true value. In addition to sampling variability, the ACS estimates are subject to nonsampling error (for a discussion of nonsampling variability, see Accuracy of the Data). The effect of nonsampling error is not represented in these tables.Data Processing Notes:This layer is updated automatically when the most current vintage of ACS data is released each year, usually in December. The layer always contains the latest available ACS 5-year estimates. It is updated annually within days of the Census Bureau's release schedule. Click here to learn more about ACS data releases.Boundaries come from the US Census TIGER geodatabases, specifically, the National Sub-State Geography Database (named tlgdb_(year)_a_us_substategeo.gdb). Boundaries are updated at the same time as the data updates (annually), and the boundary vintage appropriately matches the data vintage as specified by the Census. These are Census boundaries with water and/or coastlines erased for cartographic and mapping purposes. For census tracts, the water cutouts are derived from a subset of the 2020 Areal Hydrography boundaries offered by TIGER. Water bodies and rivers which are 50 million square meters or larger (mid to large sized water bodies) are erased from the tract level boundaries, as well as additional important features. For state and county boundaries, the water and coastlines are derived from the coastlines of the 2023 500k TIGER Cartographic Boundary Shapefiles. These are erased to more accurately portray the coastlines and Great Lakes. The original AWATER and ALAND fields are still available as attributes within the data table (units are square meters).The States layer contains 52 records - all US states, Washington D.C., and Puerto RicoCensus tracts with no population that occur in areas of water, such as oceans, are removed from this data service (Census Tracts beginning with 99).Percentages and derived counts, and associated margins of error, are calculated values (that can be identified by the "_calc_" stub in the field name), and abide by the specifications defined by the American Community Survey.Field alias names were created based on the Table Shells file available from the American Community Survey Summary File Documentation page.Negative values (e.g., -4444...) have been set to null, with the exception of -5555... which has been set to zero. These negative values exist in the raw API data to indicate the following situations:The margin of error column indicates that either no sample observations or too few sample observations were available to compute a standard error and thus the margin of error. A statistical test is not appropriate.Either no sample observations or too few sample observations were available to compute an estimate, or a ratio of medians cannot be calculated because one or both of the median estimates falls in the lowest interval or upper interval of an open-ended distribution.The median falls in the lowest interval of an open-ended distribution, or in the upper interval of an open-ended distribution. A statistical test is not appropriate.The estimate is controlled. A statistical test for sampling variability is not appropriate.The data for this geographic area cannot be displayed because the number of sample cases is too small.
In 2024, 30.69 percent of Hispanics reported speaking only English in the United States.
Open Data Commons Attribution License (ODC-By) v1.0https://www.opendatacommons.org/licenses/by/1.0/
License information was derived automatically
2013-2023 Virginia Population by Age by Language Spoken at Home by Ability to Speak English for the Population 5 years and over by Census Block Group. Contains estimates and margins of error.
U.S. Census Bureau; American Community Survey, American Community Survey 5-Year Estimates, Table B16004 Data accessed from: Census Bureau's API for American Community Survey (https://www.census.gov/data/developers/data-sets.html)
The United States Census Bureau's American Community Survey (ACS): -What is the American Community Survey? (https://www.census.gov/programs-surveys/acs/about.html) -Geography & ACS (https://www.census.gov/programs-surveys/acs/geography-acs.html) -Technical Documentation (https://www.census.gov/programs-surveys/acs/technical-documentation.html)
Supporting documentation on code lists, subject definitions, data accuracy, and statistical testing can be found on the American Community Survey website in the Technical Documentation section. (https://www.census.gov/programs-surveys/acs/technical-documentation/code-lists.html)
Sample size and data quality measures (including coverage rates, allocation rates, and response rates) can be found on the American Community Survey website in the Methodology section. (https://www.census.gov/acs/www/methodology/sample_size_and_data_quality/)
Although the American Community Survey (ACS) produces population, demographic and housing unit estimates, it is the Census Bureau's Population Estimates Program that produces and disseminates the official estimates of the population for the nation, states, counties, cities, and towns and estimates of housing units for states and counties.
Data are based on a sample and are subject to sampling variability. The degree of uncertainty for an estimate arising from sampling variability is represented through the use of a margin of error. The value shown here is the 90 percent margin of error. The margin of error can be interpreted roughly as providing a 90 percent probability that the interval defined by the estimate minus the margin of error and the estimate plus the margin of error (the lower and upper confidence bounds) contains the true value. In addition to sampling variability, the ACS estimates are subject to nonsampling error (for a discussion of nonsampling variability, see ACS Technical Documentation https://www.census.gov/programs-surveys/acs/technical-documentation.html). The effect of nonsampling error is not represented in these tables.
CC0 1.0 Universal Public Domain Dedicationhttps://creativecommons.org/publicdomain/zero/1.0/
License information was derived automatically
Although the American Community Survey (ACS) produces population, demographic and housing unit estimates, the decennial census is the official source of population totals for April 1st of each decennial year. In between censuses, the Census Bureau's Population Estimates Program produces and disseminates the official estimates of the population for the nation, states, counties, cities, and towns and estimates of housing units for states and counties..Information about the American Community Survey (ACS) can be found on the ACS website. Supporting documentation including code lists, subject definitions, data accuracy, and statistical testing, and a full list of ACS tables and table shells (without estimates) can be found on the Technical Documentation section of the ACS website.Sample size and data quality measures (including coverage rates, allocation rates, and response rates) can be found on the American Community Survey website in the Methodology section..Source: U.S. Census Bureau, 2022 American Community Survey 1-Year Estimates.Data are based on a sample and are subject to sampling variability. The degree of uncertainty for an estimate arising from sampling variability is represented through the use of a margin of error. The value shown here is the 90 percent margin of error. The margin of error can be interpreted roughly as providing a 90 percent probability that the interval defined by the estimate minus the margin of error and the estimate plus the margin of error (the lower and upper confidence bounds) contains the true value. In addition to sampling variability, the ACS estimates are subject to nonsampling error (for a discussion of nonsampling variability, see ACS Technical Documentation). The effect of nonsampling error is not represented in these tables..A "limited English speaking household" is one in which no member 14 years old and over (1) speaks only English or (2) speaks a non-English language and speaks English "very well." In other words, all members 14 years old and over have at least some difficulty with English. By definition, English-only households cannot belong to this group. Previous Census Bureau data products have referred to these households as "linguistically isolated" and "Households in which no one 14 and over speaks English only or speaks a language other than English at home and speaks English 'very well'." This table is directly comparable to tables from earlier years that used these labels..The household language assigned to the housing unit is the non-English language spoken by the first person with a non-English language in the following order: reference person, spouse, parent, sibling, child, grandchild, in-law, other relative, unmarried partner, housemate/roommate, roomer/boarder, foster child, or other nonrelative. If no member of the household age 5 and over speaks a language other than English at home then the household language is English only..In 2016, changes were made to the languages and language categories presented in tables B16001, C16001, and B16002. For more information, see: 2016 Language Data User note..The 2022 American Community Survey (ACS) data generally reflect the March 2020 Office of Management and Budget (OMB) delineations of metropolitan and micropolitan statistical areas. In certain instances the names, codes, and boundaries of the principal cities shown in ACS tables may differ from the OMB delineations due to differences in the effective dates of the geographic entities..Estimates of urban and rural populations, housing units, and characteristics reflect boundaries of urban areas defined based on 2020 Census data. As a result, data for urban and rural areas from the ACS do not necessarily reflect the results of ongoing urbanization..Explanation of Symbols:- The estimate could not be computed because there were an insufficient number of sample observations. For a ratio of medians estimate, one or both of the median estimates falls in the lowest interval or highest interval of an open-ended distribution. For a 5-year median estimate, the margin of error associated with a median was larger than the median itself.N The estimate or margin of error cannot be displayed because there were an insufficient number of sample cases in the selected geographic area. (X) The estimate or margin of error is not applicable or not available.median- The median falls in the lowest interval of an open-ended distribution (for example "2,500-")median+ The median falls in the highest interval of an open-ended distribution (for example "250,000+").** The margin of error could not be computed because there were an insufficient number of sample observations.*** The margin of error could not be computed because the median falls in the lowest interval or highest interval of an open-ended distribution.***** A margin of error is not appropriate because the corresponding estimate is controlled to an independent population or housing estimate. Effectively, the correspondi...
https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement
Welcome to the US 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 US 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 US accents and dialects.
The dataset comprises 30 hours of high-quality audio, featuring natural, free-flowing dialogue between native speakers of US Spanish. These sessions range from informal daily talks to deeper, topic-specific discussions, ensuring variability and context richness for diverse use cases.
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.
Each audio file is paired with a human-verified, verbatim transcription available in JSON format.
These transcriptions are production-ready, enabling seamless integration into ASR model pipelines or conversational AI workflows.
The dataset comes with granular metadata for both speakers and recordings:
Such metadata helps developers fine-tune model training and supports use-case-specific filtering or demographic analysis.
This dataset is a versatile resource for multiple Spanish speech and language AI applications:
Attribution 4.0 (CC BY 4.0)https://creativecommons.org/licenses/by/4.0/
License information was derived automatically
This layer was developed by the Research & Analytics Group of the Atlanta Regional Commission, using data from the U.S. Census Bureau’s American Community Survey 5-year estimates for 2013-2017, to show number and percentage of U.S. population 5 years and older that speaks English less than "very well" and don’t speak English at home by Westside Future Fund in the Atlanta region. The user should note that American Community Survey data represent estimates derived from a surveyed sample of the population, which creates some level of uncertainty, as opposed to an exact measure of the entire population (the full census count is only conducted once every 10 years and does not cover as many detailed characteristics of the population). Therefore, any measure reported by ACS should not be taken as an exact number – this is why a corresponding margin of error (MOE) is also given for ACS measures. The size of the MOE relative to its corresponding estimate value provides an indication of confidence in the accuracy of each estimate. Each MOE is expressed in the same units as its corresponding measure; for example, if the estimate value is expressed as a number, then its MOE will also be a number; if the estimate value is expressed as a percent, then its MOE will also be a percent. The user should also note that for relatively small geographic areas, such as census tracts shown here, ACS only releases combined 5-year estimates, meaning these estimates represent rolling averages of survey results that were collected over a 5-year span (in this case 2013-2017). Therefore, these data do not represent any one specific point in time or even one specific year. For geographic areas with larger populations, 3-year and 1-year estimates are also available. For further explanation of ACS estimates and margin of error, visit Census ACS website. Naming conventions: Prefixes:NoneCountpPercentrRatemMedianaMean (average)tAggregate (total)chChange in absolute terms (value in t2 - value in t1)pchPercent change ((value in t2 - value in t1) / value in t1)chpChange in percent (percent in t2 - percent in t1)Suffixes:NoneChange over two periods_eEstimate from most recent ACS_mMargin of Error from most recent ACS_00Decennial 2000 Attributes:SumLevelSummary level of geographic unit (e.g., County, Tract, NSA, NPU, DSNI, SuperDistrict, etc)GEOIDCensus tract Federal Information Processing Series (FIPS) code NAMEName of geographic unitPlanning_RegionPlanning region designation for ARC purposesAcresTotal area within the tract (in acres)SqMiTotal area within the tract (in square miles)CountyCounty identifier (combination of Federal Information Processing Series (FIPS) codes for state and county)CountyNameCounty NamePop5P_e# Population 5 years and over, 2017Pop5P_m# Population 5 years and over, 2017 (MOE)EnglishOnly_e# Speaks English only, 2017EnglishOnly_m# Speaks English only, 2017 (MOE)pEnglishOnly_e% Speaks English only, 2017pEnglishOnly_m% Speaks English only, 2017 (MOE)NotEnglish_e# Speaks language other than English at home, 2017NotEnglish_m# Speaks language other than English at home, 2017 (MOE)pNotEnglish_e% Speaks language other than English at home, 2017pNotEnglish_m% Speaks language other than English at home, 2017 (MOE)EngLtVeryWell_e# English not spoken at home, speaks English less than 'very well', 2017EngLtVeryWell_m# English not spoken at home, speaks English less than 'very well', 2017 (MOE)pEngLtVeryWell_e% English not spoken at home, speaks English less than 'very well', 2017pEngLtVeryWell_m% English not spoken at home, speaks English less than 'very well', 2017 (MOE)Spanish_e# Speaks Spanish at home, 2017Spanish_m# Speaks Spanish at home, 2017 (MOE)pSpanish_e% Speaks Spanish at home, 2017pSpanish_m% Speaks Spanish at home, 2017 (MOE)SpanishEngLtVeryWell_e# Speaks Spanish at home, speaks English less than 'very well', 2017SpanishEngLtVeryWell_m# Speaks Spanish at home, speaks English less than 'very well', 2017 (MOE)pSpanishEngLtVeryWell_e% Speaks Spanish at home, speaks English less than 'very well', 2017pSpanishEngLtVeryWell_m% Speaks Spanish at home, speaks English less than 'very well', 2017 (MOE)IndoEurNotEnglish_e# Speaks other Indo-European language at home, 2017IndoEurNotEnglish_m# Speaks other Indo-European language at home, 2017 (MOE)pIndoEurNotEnglish_e% Speaks other Indo-European language at home, 2017pIndoEurNotEnglish_m% Speaks other Indo-European language at home, 2017 (MOE)IndoEurEngLtVeryWell_e# Speaks other Indo-European language at home, speaks English less than 'very well', 2017IndoEurEngLtVeryWell_m# Speaks other Indo-European language at home, speaks English less than 'very well', 2017 (MOE)pIndoEurEngLtVeryWell_e% Speaks other Indo-European language at home, speaks English less than 'very well', 2017pIndoEurEngLtVeryWell_m% Speaks other Indo-European language at home, speaks English less than 'very well', 2017 (MOE)AsianNotEnglish_e# Speaks Asian language at home, 2017AsianNotEnglish_m# Speaks Asian language at home, 2017 (MOE)pAsianNotEnglish_e% Speaks Asian language at home, 2017pAsianNotEnglish_m% Speaks Asian language at home, 2017 (MOE)AsianEngLtVeryWell_e# Speaks Asian language at home, speaks English less than 'very well', 2017AsianEngLtVeryWell_m# Speaks Asian language at home, speaks English less than 'very well', 2017 (MOE)pAsianEngLtVeryWell_e% Speaks Asian language at home, speaks English less than 'very well', 2017pAsianEngLtVeryWell_m% Speaks Asian language at home, speaks English less than 'very well', 2017 (MOE)OthLangNotEnglish_e# Speaks other language at home, 2017OthLangNotEnglish_m# Speaks other language at home, 2017 (MOE)pOthLangNotEnglish_e% Speaks other language at home, 2017pOthLangNotEnglish_m% Speaks other language at home, 2017 (MOE)OthLangEngLtVeryWell_e# Speaks other language at home, speaks English less than 'very well', 2017OthLangEngLtVeryWell_m# Speaks other language at home, speaks English less than 'very well', 2017 (MOE)pOthLangEngLtVeryWell_e% Speaks other language at home, speaks English less than 'very well', 2017pOthLangEngLtVeryWell_m% Speaks other language at home, speaks English less than 'very well', 2017 (MOE)last_edited_dateLast date the feature was edited by ARC Source: U.S. Census Bureau, Atlanta Regional CommissionDate: 2013-2017 For additional information, please visit the Census ACS website.
https://www.futurebeeai.com/policies/ai-data-license-agreementhttps://www.futurebeeai.com/policies/ai-data-license-agreement
This US Spanish Call Center Speech Dataset for the BFSI (Banking, Financial Services, and Insurance) sector is purpose-built to accelerate the development of speech recognition, spoken language understanding, and conversational AI systems tailored for Spanish-speaking customers. Featuring over 30 hours of real-world, unscripted audio, it offers authentic customer-agent interactions across a range of BFSI services to train robust and domain-aware ASR models.
Curated by FutureBeeAI, this dataset empowers voice AI developers, financial technology teams, and NLP researchers to build high-accuracy, production-ready models across BFSI customer service scenarios.
The dataset contains 30 hours of dual-channel call center recordings between native US Spanish speakers. Captured in realistic financial support settings, these conversations span diverse BFSI topics from loan enquiries and card disputes to insurance claims and investment options, providing deep contextual coverage for model training and evaluation.
This speech corpus includes both inbound and outbound calls with varied conversational outcomes like positive, negative, and neutral, ensuring real-world BFSI voice coverage.
This variety ensures models trained on the dataset are equipped to handle complex financial dialogues with contextual accuracy.
All audio files are accompanied by manually curated, time-coded verbatim transcriptions in JSON format.
These transcriptions are production-ready, making financial domain model training faster and more accurate.
Rich metadata is available for each participant and conversation:
Mexico is the country with the largest number of native Spanish speakers in the world. As of 2024, 132.5 million people in Mexico spoke Spanish with a native command of the language. Colombia was the nation with the second-highest number of native Spanish speakers, at around 52.7 million. Spain came in third, with 48 million, and Argentina fourth, with 46 million. Spanish, a world language As of 2023, Spanish ranked as the fourth most spoken language in the world, only behind English, Chinese, and Hindi, with over half a billion speakers. Spanish is the official language of over 20 countries, the majority on the American continent, nonetheless, it's also one of the official languages of Equatorial Guinea in Africa. Other countries have a strong influence, like the United States, Morocco, or Brazil, countries included in the list of non-Hispanic countries with the highest number of Spanish speakers. The second most spoken language in the U.S. In the most recent data, Spanish ranked as the language, other than English, with the highest number of speakers, with 12 times more speakers as the second place. Which comes to no surprise following the long history of migrations from Latin American countries to the Northern country. Moreover, only during the fiscal year 2022. 5 out of the top 10 countries of origin of naturalized people in the U.S. came from Spanish-speaking countries.