Audio Transcriptions API
Reference for POST /v1/audio/transcriptions, including multipart fields, supported transcription models, token pricing, minimum billing, and error responses.
Endpoint
POST https://api.undersky.ai/v1/audio/transcriptionsAuthenticate with a model API key in the Authorization header. Send the request as multipart/form-data.
Form fields
| Field | Type | Required | Description |
|---|---|---|---|
file | file | Yes | Audio file with a filename. |
model | string | No | whisper-1, gpt-4o-mini-transcribe, gpt-4o-transcribe, or gpt-4o-transcribe-diarize. Defaults to gpt-4o-mini-transcribe. |
response_format | string | No | Response format supported by the configured upstream. |
language | string | No | ISO-639-1 language hint. |
prompt | string | No | Vocabulary or context hint for the transcription. |
temperature | number | No | Sampling temperature supported by the upstream. |
Other multipart fields are forwarded to the configured transcription provider.
Request
curl -sS https://api.undersky.ai/v1/audio/transcriptions \
-H "Authorization: Bearer sk-your-api-key" \
-F "[email protected]" \
-F "model=gpt-4o-mini-transcribe" \
-F "language=en" \
-F "response_format=json"Models and pricing
| Model | Input price | Output price |
|---|---|---|
whisper-1 | fixed request pricing | fixed request pricing |
gpt-4o-mini-transcribe | $1 / 1M tokens | $4 / 1M tokens |
gpt-4o-transcribe | $84 / 1M tokens | $140 / 1M tokens |
gpt-4o-transcribe-diarize | $84 / 1M tokens | $140 / 1M tokens |
Successful requests have a $0.10 credit minimum. Failed requests release the billing hold.
Responses
The gateway returns the successful JSON or text response from the configured provider and preserves its content type. Keep response parsing compatible with the selected model and format.
Errors
| Status | Code | Meaning |
|---|---|---|
400 | invalid_multipart or unsupported_model | The form body, file, or model is invalid. |
401 | authentication error | The API key is missing or invalid. |
402 | billing error | The account cannot reserve the minimum credit amount. |
429 | rate limit | Retry with bounded exponential backoff. |
502 | upstream_unreachable | The configured provider did not respond. |
See the transcription guide for a complete workflow and the media API overview for related endpoints.
Text to Speech API
Complete reference for OpenAI-compatible speech synthesis through Tuzi API, including supported models, voices, formats, speed controls, authentication, pricing, and binary audio responses.
Image Generation (Create)
Reference for POST /v1/images/generations, including prompts, sizes, reference images, async task creation, credits, and callbacks.