Audio Transcription

POST /v1/audio/transcriptions

Transcribe audio files to text. Compatible with OpenAI’s Whisper API.

Multipart Request

curl -X POST "$BASE/v1/audio/transcriptions" \
  -H "Authorization: Bearer $TOKEN" \
  -F "model=openai/whisper-1" \
  -F "durationSeconds=12.3" \
  -F "file=@/path/to/audio.webm;type=audio/webm"

JSON Request

{
  "model": "openai/whisper-1",
  "durationSeconds": 12.3,
  "fileBase64": "data:audio/webm;base64,..."
}

Supported Providers

• OpenAI Whisper - General-purpose transcription
• Azure Speech - With pronunciation assessment support

Performance tip: For the fastest unscripted pronunciation-assessment latency, send WAV containing 16 kHz mono 16-bit PCM. Other formats (WebM/Opus from browser MediaRecorder, MP3, M4A, FLAC, OGG, AAC) are fully supported but require a ~30–100 ms server-side transcode.

For live streaming transcription over WebSocket, see Realtime Audio - Live Transcription.