Muse Voice Transcribe 1.0
meta/muse-voice-transcribe-1.0Transcribe a WAV recording.
- second
- —
- Meta
- 5 min
Overview
File transcription in PUSH_TO_TALK mode: accepts mono 16-bit PCM WAV at 16 or 24 kHz, up to 10 minutes with a 23 MiB decoded-file limit and a 31 MiB JSON input limit. Ten minutes is supported at 16 kHz; at 24 kHz the file-size limit permits about 8 minutes 22 seconds for a standard 44-byte WAV header. Base64 expands the upload, so the complete HTTP request must remain below the 32 MiB public ingress limit. Billing uses supplier-reported processed audio duration rounded down to whole seconds, with no minimum. The supplier standard rate is $0.18 per hour ($0.00005 per whole second); customer pricing follows the account price plan. ENDPOINTING, DIARIZATION and realtime streaming are not available through this endpoint.
The transcript, sessionId and audioDurationMs. Duration is measured from the response without a caller-provided override.
Input
These go in the input object of the run request.
| Field | Type | Required | Description |
|---|---|---|---|
| audio_base64 | base64 | Yes | A complete mono 16-bit PCM WAV recording, base64-encoded. |
| mode | string | No | Only PUSH_TO_TALK is enabled; other native modes require separate acceptance. |
| keywords | array | No | Strings biasing recognition toward names or terms. |
| languageBias | array | No | Strings hinting at expected languages. |
Requests up to 33 MB, audio up to 24 MB.
Example
A call that works, with your own key.
curl https://gateway.agentsky.dev/v1/run \
-H "authorization: Bearer $AGENTSKY_API_KEY" \
-H "content-type: application/json" \
-d '{"provider":"meta","endpoint":"muse-voice-transcribe-1.0","input":{"audio_base64":"UklGRiYAAABXQVZFZm10IBAAAAABAAEAgD4AAAB9AAACABAAZGF0YQIAAAAAAA==","mode":"PUSH_TO_TALK"}}'Pricing
— per second on the default plan. Every run carries its price in the response.