APIリファレンス · Media
Transcription
Transcribe an audio file to text.
POST/v1/audio/transcriptions1 per started minute of audio
Send multipart/form-data with a file part — wav or mp3 (other containers return unsupported_format; re-encode client-side or record WAV via WebAudio). Language is detected automatically; pass language (ISO code) as a hint to improve accuracy.
Billed per started minute of audio, 1 credit each — a 20-second voice note is 1 credit, a 2.5-minute recording 3. The length is read from the file header before transcription and returned as usage.audio_seconds; a failed transcription refunds itself.
Files are capped at 8MB (about 8 minutes of 128 kbps mp3, far more at voice bitrates).
リクエストボディ
filefile必須
The audio to transcribe (wav or mp3).
modelstring
Only eroq-scribe-one today.
languagestring
ISO language hint, e.g. en, fr.
curl https://eroq.ai/v1/audio/transcriptions \
-H "Authorization: Bearer $EROQ_API_KEY" \
-F model=eroq-scribe-one \
-F [email protected]/v1/audio/transcriptions
→ POST /v1/audio/transcriptions
「例を実行」を押すと、ドキュメント自身のデータから実際のやり取りを再生します。キーもリクエストも不要で、料金もかかりません。
レスポンス
200 · application/json
{
"model": "eroq-scribe-one",
"text": "That noise is the reactor missing its coolant flush.",
"usage": { "audio_seconds": 4.2, "credits_spent": 1, "credits_remaining": 882 }
}