Japanese speech-to-text API (日本語)

Transcribe Japanese with one request

Set language to ja and start a job. Poll GET /transcribe/{jobId} for the result, or pass a webhook_url.

curl https://app.kolwrite.com/transcribe \
  -H "x-api-key: $KOLWRITE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"audio_url": "https://example.com/meeting-ja.m4a", "model": "oracle", "language": "ja", "include_diarization": true, "output_format": "json"}'

Translate Japanese audio into English in the same job (Oracle models only):

curl https://app.kolwrite.com/transcribe \
  -H "x-api-key: $KOLWRITE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"audio_url": "https://example.com/meeting-ja.m4a", "model": "oracle_mini", "language": "ja", "translate_to": "en", "output_format": "srt"}'

Japanese at a glance

What to know about Japanese transcription

• Japanese has no spaces between words, so word timestamps may cover single characters or short tokens. Cut subtitles by time and by character count.
• The same word can be written in kanji, hiragana or katakana. If you search transcripts, normalize to one form (for example with a morphological analyzer).
• Business Japanese uses honorific forms and many English loanwords in katakana; send ja explicitly on mixed recordings.

Where teams use Japanese transcription

• Meeting and interview transcripts for teams working with Japan.
• Japanese subtitles and translated subtitles for video.
• User research in Japan, translated with translate_to: "en".

Choosing a model

Start with oracle (US$1.44 per hour) for production Japanese audio. Use base (US$0.48) for clean, single-speaker recordings at volume, oracle_mini (US$0.96) for the cheapest translation, and oracle_pro or oracle_max for noisy or difficult audio. KolWrite does not publish a Japanese accuracy figure for this API, so test a sample of your own recordings before committing. Full comparison: models and prices.

Questions and answers