Kotoba STT v1
kotoba-technologies/kotoba-stt/v1/speech-to-textTranscribe audio or video clips up to 120 seconds in Japanese, English, Korean, Mandarin, Taiwanese Mandarin, Hong Kong Cantonese, or Spanish. Requires an explicit spoken language; no language detection or translation. Optional word timestamps and beta speaker diarization.
- Speech to text
Pricing
Includes our margin. Charged per request from your balance.
- Per minute
- $0.0078
Details
- Provider
- Kotoba-technologies
- Added
- Nov 5, 2026
Options
Parameters this model accepts.
- Languages
- ja
- en
- ko
- zh
- zh-tw
- zh-hk
- es
- Features
- Audio and video clips up to 120 seconds, Word-level timestamps, Beta speaker diarization, Explicit spoken language
- Response formats
- json
- text
- verbose_json
- Input formats
- WAV, MP3, M4A, MP4, WEBM, OGG, FLAC, AAC
- Max file size
- 100 MB
- Max upload in the request
- 4 MB
- File URLs
- Supported
- Timestamps
- word