OpenAISpeech to textComing soon

gpt-transcribe

Name: gpt-transcribe
Brand: OpenAI

gpt-transcribe is a speech to text tool from OpenAI that runs in the background and returns structured JSON. Coming soon: listed with its price, not yet open to runs.

This tool is listed but not yet available to run.

Price pending

Pricing

Price pending.

Limits

Runs: In the background
Returns: json
Longest queue wait: 3600s
Run deadline: 1800s
Largest inline input: 1024 KB
Largest input file: 500 MB
Largest output: 4768 MB
Outputs per run: 16

Input

Field	Kind	Default	Description
`file` required	URL or file	—	The audio file object (not file name) to transcribe, in one of these formats: flac, mp3, mp4, mpeg, mpga, m4a, ogg, wav, or webm. The request must include enough format metadata for the file to be identified. We recommend an extension-bearing filename and an appropriate content type.
`prompt`	text	—	An optional text to guide the model's style or continue a previous audio segment. The [prompt](https://developers.openai.com/api/docs/guides/speech-to-text#prompting) should match the audio language. This field is not supported when using `gpt-4o-transcribe-diarize`.
`include`	list	—	Additional information to include in the transcription response. `logprobs` will return the log probabilities of the tokens in the response to understand the model's confidence in the transcription. `logprobs` only works with response_format set to `json` and only with the models `gpt-4o-transcribe`, `gpt-4o-mini-transcribe`, and `gpt-4o-mini-transcribe-2025-12-15`. This field is not supported when using `gpt-4o-transcribe-diarize`.
`keywords`	list	—	Words or phrases to guide transcription of the input audio. Supported by `gpt-transcribe`.
`language`	text	—	The language of the input audio. Supplying the input language in [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) (e.g. `en`) format will improve accuracy and latency.
`languages`	list	—	Possible languages of the input audio, in [ISO-639-1](https://en.wikipedia.org/wiki/List_of_ISO_639-1_codes) format. Supported by `gpt-transcribe`.
`temperature`	number	0	The sampling temperature, between 0 and 1. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. If set to 0, the model will use [log probability](https://en.wikipedia.org/wiki/Log_probability) to automatically increase the temperature until certain thresholds are hit.
`response_format`	one of json, text, srt, verbose_json, vtt, diarized_json	json	The format of the output, in one of these options: `json`, `text`, `srt`, `verbose_json`, `vtt`, or `diarized_json`. For `gpt-4o-transcribe` and `gpt-4o-mini-transcribe`, the only supported format is `json`. For `gpt-4o-transcribe-diarize`, the supported formats are `json`, `text`, and `diarized_json`, with `diarized_json` required to receive speaker annotations.
`chunking_strategy`	JSON	—	Controls how the audio is cut into chunks. When set to `"auto"`, the server first normalizes loudness and then uses voice activity detection (VAD) to choose boundaries. `server_vad` object can be provided to tweak VAD detection parameters manually. If unset, the audio is transcribed as a single block. Required when using `gpt-4o-transcribe-diarize` for inputs longer than 30 seconds.
`known_speaker_names`	list	—	Optional list of speaker names that correspond to the audio samples provided in `known_speaker_references[]`. Each entry should be a short identifier (for example `customer` or `agent`). Up to 4 speakers are supported.
`timestamp_granularities`	list	["segment"]	The timestamp granularities to populate for this transcription. `response_format` must be set `verbose_json` to use timestamp granularities. Either or both of these options are supported: `word`, or `segment`. Note: There is no additional latency for segment timestamps, but generating word timestamps incurs additional latency. This option is not available for `gpt-4o-transcribe-diarize`.
`known_speaker_references`	list	—	Optional list of audio samples (as [data URLs](https://developer.mozilla.org/en-US/docs/Web/HTTP/Basics_of_HTTP/Data_URLs)) that contain known speaker references matching `known_speaker_names[]`. Each sample must be between 2 and 10 seconds, and can use any of the same input audio formats supported by `file`.

Same request in code

curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "openai",
    "model": "gpt-transcribe",
    "input": {
      "file": "<file>"
    }
  }'