POSThttps://api.momor.ai/models/transcribe
Here auto, *, a tier and a name all reach transcription models alone. For OpenAI’s multipart shape, use Create a transcription.
Body
The body is the one Call a model takes. A transcription takes no streaming, tools or responseSchema.
- systemPromptstringRequired
- Words the recording uses, passed on to the model. May be empty.
- messagesMessage[]Required
- One user message carrying the recording as its one file.
- typestringRequired
- user
- valuestring
- May be empty.
- filesFile[]Required
- Exactly one audio/* file, sent inline as bytes. A recording sent by address is refused.
- modelstring | objectDefault
auto - Which transcription models may answer, as on Call a model.
- models(string | object)[]
- Several selectors, tried in order.
- prioritizationobject
- Which providers lead and which serve nothing, and zdr, zero data retention, as on Call a model.
- keysobject
- Your own provider keys for this call.
- parametersobject
- Sampling, as on Call a model.
- temperaturenumber
- Passed on to the model.
Response
- typestring
- monolith
- modelobject
- What served the call: name, tier, publisher and provider.
- messagesMessage[]
- One content message whose value is the transcript.
- usageobject
- What the provider counted the recording as. A token count the provider does not report is left at 0.
- inputTokensinteger
- Input tokens, for a model priced by the token.
- outputTokensinteger
- Output tokens, for a model priced by the token.
- unitsobject
- The length of the recording, where the provider reports one: unit is minute and count the minutes, in fractions of a minute. A model priced by the minute is charged count times its perUnit price. Left out where the provider reports no length.
Errors
Failures are problem documents, as on Call a model. The messages particular to a transcription: