Japanese live conversational audio, 4-channel multi-mic, transcribed
Overview
Japanese live conversational audio, 4-channel multi-mic, transcribed is an audio dataset. It covers Japanese (Japan). The provider states a volume of 106 hours. Typical applications include speech recognition and conversational AI. The listing mentions transcription. Pricing is not published and is provided on request.
Primary applications
What the listing offers
- Stated volume of 106 hours.
- Regional coverage: Japanese (Japan).
- Listing mentions transcription.
- A sample is indicated as available from the provider.
A collection of various audio including retail, telecom, real estate, finance, medical, etc. with time aligned transcript, fillers, full annotation, and metadata recorded in studio including 4-channel multi-mic. in addition to the studio recording, this includes the audio tracks from the other three microphones: gooseneck, lavalier, and smartphone.
Technical specifications
- Modalities
Related datasets
More in Speech & Audio →Japanese Conversational Speech by Mobile Phone | 633 Hours | Multilingual Language Data
Japanese(Japan) Spontaneous Dialogue Smartphone speech dataset, collected from dialogues based on given topics.
- Language
- Japanese
- Volume
- 633 h
- Domain
- AI & Machine Learning +1
- Use case
- Speech recognition
More Japanese speech & audio datasets
All 55 Japanese speech & audio datasets →- Japanese live conversational audio, channel-separated
- Japanese live conversational audio, single-channel
- Japanese live conversational audio, transcribed, annotated, remote-recorded
- Japanese live conversational audio, transcribed, sound-isolated booths
- Japanese Multi-Turn Speech Dataset for Everyday, Travel & Academic Domains
- Japanese Natural Dialogue Speech Dataset
- Japanese podcast audio, studio-recorded
- Japanese podcast audio, studio-recorded
- Japanese podcast audio, studio-recorded
- Japanese retail IVR audio, telephony