en

Please fill in your name

Mobile phone format error

Please enter the telephone

Please enter your company name

Please enter your company email

Please enter the data requirement

Successful submission! Thank you for your support.

Format error, Please fill in again

Confirm

The data requirement cannot be less than 5 words and cannot be pure numbers

m.nexdata.datatang.com

Home > All Category Datasets > Speech Recognition Datasets > 357 Hours Korean Speech Dataset with 999 Native Speakers for AI Training

357 Hours Korean Speech Dataset with 999 Native Speakers for AI Training

korean audio dataset

korean speech recognition dataset

korean speech dataset

korean ASR dataset

korean speech corpus

This dataset contains 357 hours of Korean scripted speech recordings collected from 999 native Korean speakers based on predefined prompts. The dataset covers general speech scenarios and provides high-quality transcriptions with additional metadata attributes. 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.

This is a paid datasets for commercial use, research purpose and more. Licensed ready made datasets help jump-start AI projects.

Specifications

Specifications

Format

16kHz, 16bit, wav, mono channel;

Recording condition

Low background noise(indoor);

Content category

Generic domain;

Recording device

Android Smartphone, iPhone;

Speaker

999 Korean, 47% male and 53% female;

Country

South Korea(KOR);

Language(Region) Code

ko-KR;

Language

Korean;

Features of annotation

Transcription text;

Accuracy Rate

Sentence Accuracy Rate (SAR) 95%

Sample

Sample

Audio
실제로 온라인데이팅 서비스를 이용하는 이들조차 그 사실을 숨김
Audio
그리고 적합한 위치에 있는지를 확인하기 위해서 호스를 잘 쳐다보아야 한다
Audio
게임의 목적은 종점에 도착함으로써 스테이지를 클리어하는 것이다
Audio
거기에 고마쓰 씨의 고백 시어터 란 제목의 고백 메세지를 보내면
Audio
클레어가 클레이모어가 되기 전 세대의 전사들

Recommended Datasets

Recommended Dataset

240 Hours Hindi Speech Dataset for ASR and AI Model Training

This dataset contains 240 hours of Hindi speech recordings collected from 401 native Indian speakers through scripted monologue tasks based on predefined scripts. The dataset covers diverse speech domains, including economy, entertainment, news, informal expressions, numbers, alphabet reading, and general language scenarios. Recordings were collected from both quiet and noisy conditions to capture real-world acoustic variations. Each audio sample is provided with accurate transcription and speaker-related metadata. 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.

hindi speech corpus hindi audio dataset hindi asr dataset hindi conversational speech dataset hindi speech dataset

227 Hours Multi-Accent Spanish Speech Dataset with Transcripts for ASR Training

This dataset contains 227 hours of Spanish scripted speech recorded as monologues based on predefined texts. It includes recordings from 352 native speakers from Spain, Mexico, Venezuela, and other Spanish-speaking regions. The speech content covers economics, entertainment, news, informal language, numbers, alphabet sequences, and other general domains. Each recording includes high-quality transcripts, timestamps, noise labels, and additional speaker metadata. 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.

spanish speech to text dataset spanish asr dataset spanish speech recognition dataset spanish speech dataset for asr training smartphone speech dataset

231.9 Hours French Speech Dataset with Diverse Speakers for AI Training

This dataset contains 231.9 hours of French scripted monologue speech collected from native speakers based on predefined scripts. The recordings cover multiple domains, including economy, entertainment, news, informal speech, numbers, and alphabet sequences. Each audio sample includes accurate transcripts and additional metadata. The dataset was collected from 406 speakers from France, Canada, and African French-speaking regions, 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.

french speech dataset french audio dataset french asr dataset european speech dataset voice assistant dataset

199 Hours British English Speech Dataset for ASR and Voice AI Training

This dataset contains 199 hours of scripted British English speech collected from 346 native British speakers using smartphone devices. The recordings are based on predefined texts covering various domains, including news, entertainment, economy, informal expressions, numbers, alphabetic content, and other general speech scenarios. Each audio sample is transcribed with corresponding text content and additional metadata attributes. 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.

british english speech dataset uk english speech dataset english ASR dataset read speech dataset speech recognition training data native british speech dataset tts training data

215 Hours American English Speech Dataset with Reading-style Speech

This dataset contains 215 hours of American English scripted read speech collected from 349 native speakers using mobile devices. The recordings are based on predefined scripts covering various content categories, including economy, entertainment, news, informal speech, numbers, and alphabet reading. Each audio sample is transcribed with corresponding text content and additional metadata. 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.

us english speech dataset english ASR dataset american english speech dataset speech AI dataset multilingual speech dataset

127 Hours Malay Speech Dataset with 156 Native Speakers for AI Training

This dataset contains 127 hours of Malay speech recordings collected from 156 native Malaysian speakers through scripted monologue tasks based on predefined texts. The dataset covers diverse speech domains, including economy, entertainment, news, informal expressions, numbers, alphabet reading, and general language scenarios. Each audio sample is provided with accurate transcription and additional metadata attributes. Collected from diverse speakers with different speaking styles, this dataset can enhance 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.

malay speech dataset malay ASR dataset malay speech recognition dataset native malay speech data malay speech corpus

359 Hours Indonesian Speech Dataset for ASR and Speech Recognition Training

This dataset contains 359 hours of scripted Indonesian speech recordings collected from 496 native Indonesian speakers based on predefined scripts. The recordings cover diverse content domains, including economy, entertainment, news, informal expressions, numbers, and alphabet-related speech. Each audio sample is provided with accurate transcription and additional speaker metadata attributes. 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.

indonesian speech dataset bahasa indonesia speech dataset indonesian audio dataset indonesian ASR dataset

203 Hours Thai ASR Dataset with Prompt-based Speech Recordings

This dataset contains 203 hours of Thai speech collected through prompted monologue recordings using smartphone devices. The recordings are based on predefined scripts and cover multiple domains, including economics, entertainment, news, conversational language, numbers, and letters. Each audio includes accurate text transcripts and related metadata. The dataset was collected from 498 native Thai speakers with diverse demographic and geographic 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.

Thai Speech Dataset Thai ASR Dataset Thai Speech Recognition Dataset Thai Audio Dataset

Tell Us Your Special Needs

Current Project Maturity

Early exploration (no concrete specs yet)

Defined goals, need professional guidance

Active development or optimization phase

Data & labeling experts with clear specifications

Full Name *

Contact Phone No.*

Company name *

Company Email *

Data Requirements *

By submitting, I agree to the Privacy Protection

What languages and scenarios are covered by Nexdata’s speech recognition datasets?

Nexdata offers speech recognition datasets covering a broad range of languages, dialects, and accents, backed by extensive global language resources. Our datasets include diverse speakers, acoustic environments, and real-world speech scenarios, supporting multilingual ASR, voice assistants, conversational AI, speech-to-text, and other speech-enabled applications.

Can Nexdata customize speech recognition datasets for specific languages or requirements?

Yes. If our off-the-shelf datasets do not fully meet your requirements, Nexdata provides flexible custom data collection, transcription, and annotation services. We can customize datasets based on target languages or dialects, speaker profiles, recording environments, speech scenarios, data volume, and annotation specifications to meet specific ASR development needs.

How does Nexdata ensure the quality and scalability of speech recognition datasets?

Nexdata applies multi-stage quality control throughout speech data collection, transcription, annotation, and validation. Combined with our extensive language resources and scalable collection capabilities, we can support both large-scale multilingual projects and specialized datasets for specific languages, dialects, accents, and speech scenarios.

Subscribe to our newsletter

Be the first to receive Nexdata latest product releases, data solutions and enterprise news.

Off-the-Shelf Datasets: All Category Datasets; Embodied AI Datasets; LLM Datasets; Computer Vision Datasets; Speech Recognition Datasets; Speech Synthesis Datasets; OCR Datasets; Pronunciation Dictionary; NLU Datasets

Data Service: 3D Point Cloud Data; Street View Data; OCR Data; Behavior Recognition Data; Identity Recognition Data; Speech Recognition Data; Speech Synthesis Data; Multimodal Data

Industries: Embodied AI; Generative AI; Autonomous Vehicles; AR/VR; Conversational AI; Smart Home; Retail; Intelligent Healthcare

Company: About Us; News; Partners; Quality & Security; Event
Links: OPENMPD; DataPlus; Datarade

Platform: Platform
Competition: Competition
Resources: Sponsored Datasets

Sharpen Your AI with Better Data

+1(626)594-5598

[email protected]

nexdata_ai facebook

nexdata_ai twitter

nexdata_ai linkedin

nexdata_ai youtube

Copyright © 2023 NEXDATA TECHNOLOGY INC

Sitemap Terms and Conditions

We use cookies to enhance your browsing experience, serve personalized ads or content, and analyze our traffic. By clicking "Accept All", you consent to our use of cookies.

6a16ca19-349b-4cf7-a751-64709c318574

8ceaaebd-b776-4467-bf9e-deb3458ee3d6