Audio · Multimodal / Education
10,000+ Hrs Korean Conversational Speech & Transcript Data for Machine Learning (ML) Data | Multi-Speaker, Speaker-Tagged | Built for ASR & Voice AI
Advertised by providerLicensing rights not independently verifiedPrice on request
Overview
10,000+ Hrs Korean Conversational Speech & Transcript Data for Machine Learning (ML) Data | Multi-Speaker, Speaker-Tagged | Built for ASR & Voice AI is a multimodal dataset. It covers Korean. Typical applications include speech recognition, speaker recognition and conversational AI. Relevant sectors: Education and AI & Machine Learning. The listing mentions transcription, timestamps and speaker labels. Pricing is not published and is provided on request.
Primary applications
What the listing offers
- Listing mentions transcription, timestamps, speaker labels.
Specifications: ‣ Volume: 100,000+ hours ‣ Format: 8kHz, mono channel (WAV audio + JSON transcript) ‣ Environment: Studio, podcast-quality recording ‣ Recording content: Fully natural conversation, topic-guided across education, entertainment, food, legal, and automotive domains ‣ Annotation: Speaker diarization, speaker ID, gender tags, timestamps, full transcription ‣ Accuracy: 98%+ word accuracy ‣ Language: Mandarin Chinese, English, Portuguese, Spanish, and more ‣ Application scenarios: Speech recognition; conversational AI; multilingual NLP ‣ Availability: Shelf-ready; additional volume, languages, or domain coverage available on request About the provider Data: 8+ years producing AI training data, machine learning (ML) data, with 100+ in-house specialists across offices in Canada and China.
Technical specifications
Related datasets
More in Multimodal & Agent Training →Audio · MultimodalAdvertised
10,000+ Hours English Conversational Audio | Multi-Speaker, Diarized Transcripts | Speech Recognition & AI Training | WAV + JSON
Specifications: • Language: Mandarin Chinese, English, Portuguese, Spanish, and more • Volume: 100,000+ hours • Format: 8kHz, mono channel (WAV audio + JSON transcript)…
- Language
- English
- Volume
- 10k h
- Domain
- Education +1
- Use case
- Speech recognition
Price on request
More Korean speech & audio datasets
All 38 Korean speech & audio datasets →- Code-Switching Multi-Turn Speech Dataset
- Emergency Voice Command Dataset
- English-Korean Parallel Speech Dataset
- Korean Average Tone Speech Synthesis Corpus | 2 People | Text-to-Speech(TTS) Data
- Korean banking call-center audio, telephony
- Korean Call Center Consultation Speech Dataset with Emotion Labels
- Korean conversational audio, channel-separated speakers
- Korean conversational podcasts across genres
- Korean conversational podcasts spanning education, religion, culture
- Korean conversational podcasts, various genres