File transcription
Upload a WAV audio file and return a transcript with Muse Voice Transcribe.
Upload a complete WAV audio file and use Muse Voice Transcribe to receive a JSON, SSE, or plain-text transcript. The file must use mono, 16-bit integer PCM audio sampled at 16 kHz or 24 kHz.
curl --fail-with-body 'https://api.agtcloud.cn/v1/asr/transcribe' \
-H 'Authorization: Bearer your-api-key' \
-H 'Accept: application/json' \
-F 'request={"model":"muse-voice-transcribe-1.0","audioEncoding":"WAV","mode":"DIARIZATION","languageBias":["Mandarin Chinese"]};type=application/json' \
-F 'audio=@audio.wav;type=audio/wav'/v1/asr/transcribeSubmit JSON configuration and WAV audio together as multipart form data. The audio must be mono, 16-bit integer PCM WAV sampled at 16 kHz or 24 kHz, no longer than 600 seconds, and the complete request body must not exceed 32 MiB.
Use Accept to select a JSON, SSE, or plain-text response. Model availability is determined by the current API key and model catalog.
Authorization
bearerAuth Use Bearer your-api-key in the Authorization header.
In: header
Request Body
multipart/form-data
Send JSON in the request part and upload the WAV file in the audio part. The client should generate the multipart boundary automatically.
TypeScript Definitions
Use the request body type in TypeScript.