Capability

Transcription

Transcribe audio to text, with optional word-level timestamps.

transcribe — native payload
Response
Pricing

Native only — new capability

No compat route exists for transcription — it wasn’t part of the old Enterprise API this compat layer mirrors. Call POST /v1/generations directly.

transcribe — native payload

POST/v1/generationstool_id for (model="whisper-transcribe-v1", "transcribe")

Field	Type	Required	Default	Description
song	file input	Yes	—	The audio to transcribe. Whisper accepts one file up to ~25MB; a larger master is auto-compacted to fit.
language	string	No	""	Optional ISO 639-1 code, e.g. 'en', 'hi', 'ko'. Leave blank to auto-detect — forcing a wrong code produces garbled output.
word_timestamps	boolean	No	true	Return per-word start/end timestamps in addition to the transcript text.
force	boolean	No	false	Re-transcribe even if a cached transcript already exists for this file.

curl -X POST https://apiv2.soundverse.ai/v1/generations \
  -H "Authorization: Bearer sksoundverse_..." \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: transcribe-req-001" \
  -d '{
    "tool_id": "<whisper-transcribe-v1 tool_id>",
    "payload_json": "{\"song\": {\"url\": \"https://example.com/track.mp3\"}, \"word_timestamps\": true}"
  }'

Response

Once the task completes, its output has this shape:

{
  "text": "full transcript...",
  "language": "en",
  "words": [{"word": "hello", "start": 0.0, "end": 0.32}, ...],
  "segments": [{"text": "hello there", "start": 0.0, "end": 1.2}, ...]
}

words is only populated when word_timestamps was true.

Pricing

Version	Royalty-Free	Standard	Distribution	Sync	Master
whisper-transcribe-v1	$0.03	$0.03	$0.03	$0.03	$0.03

USD per transcription, by license tier.

← LyricsSound Effects →