Audio transcriptions (STT)

Transcribe audio to text. OpenAI-compatible.

POST https://api.ecohash.com/v1/audio/transcriptions

Headers

Header	Value
Authorization	Bearer eco_YOUR_KEY
Content-Type	multipart/form-data (set by your client, not by you)

Request form fields

This endpoint takes multipart/form-data, not JSON.

Field	Type	Required	Notes
model	string	yes	STT model ID (e.g. large-v3)
file	file	yes	Audio — mp3, wav, m4a, webm, ogg, flac
language	string	no	ISO-639-1 code (en, es, zh, etc.). Leave blank to auto-detect
prompt	string	no	Context hint to bias transcription (domain terms, names)
response_format	string	no	"json" (default), "text", "srt", "vtt", "verbose_json"
temperature	number	no	Sampling temperature, default 0

Example request

curl https://api.ecohash.com/v1/audio/transcriptions \
  -H "Authorization: Bearer eco_YOUR_KEY" \
  -F model=large-v3 \
  -F file=@recording.mp3

Response (default: json)

{ "text": "Full transcript here." }

Response (verbose_json)

{
  "task": "transcribe",
  "language": "en",
  "duration": 8.5,
  "text": "Full transcript.",
  "segments": [
    { "start": 0.0, "end": 4.0, "text": "First segment." },
    { "start": 4.0, "end": 8.5, "text": "Second segment." }
  ]
}

Limits

• Max file size: 25 MB
• Max duration: 30 minutes per request

For longer audio, split client-side into 5–10 minute chunks and concatenate results.

Errors

See Errors. Common: 400 (unsupported audio format), 401, 402, 404, 429.