Speech-to-Text

Transcribe audio using Whisper.

Endpoint

POST /v2/audio/asr/whisper

Request

{
  "audio_url": "https://example.com/audio.wav",
  "task": "transcribe",
  "language": "en"
}

Parameter	Type	Description
audio_url	string	URL to audio file (required)
task	string	transcribe or translate (default: transcribe)
language	string	ISO 639-1 code (auto-detected if omitted)
return_timestamps	bool	Include word timestamps

Response

{
  "text": "Hello world, this is a test.",
  "language": "en",
  "duration": 3.5,
  "task_address": "direct:abc12345-..."
}

Field	Description
text	Transcribed text
language	Detected or specified language
duration	Audio duration in seconds
task_address	Unique identifier for this request

Example

curl -X POST https://relay.opengpu.network/v2/audio/asr/whisper \
  -H "X-API-Key: YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "audio_url": "https://example.com/speech.mp3",
    "task": "transcribe"
  }'

Translation

Translate audio to English:

{
  "audio_url": "https://example.com/french-audio.wav",
  "task": "translate"
}