OpenAI

The Audio API supports two speech to text endpoints:

- <code>transcriptions</code>
- <code>transalations</code>

To get started with the Audio API, please read our <a href="https://platform.openai.com/docs/guides/speech-to-text" rel="nofollow noopener noreferrer" target="_blank">speech to text developer documentation</a>.

How much does the Audio API cost to use?

See our <a href="https://platform.openai.com/docs/pricing" rel="nofollow noopener noreferrer" target="_blank">pricing page</a> for details.

View a list of <a href="https://platform.openai.com/docs/guides/speech-to-text/supported-languages/#supported-languages" rel="nofollow noopener noreferrer" target="_blank">supported languages here</a>.

How can we handle large audio files?

The maximum file size for the Audio API is 25MB. If you expect users to upload audio files larger than 25MB, you can <a href="https://platform.openai.com/docs/guides/speech-to-text/longer-inputs" rel="nofollow noopener noreferrer" target="_blank">follow our documentation</a> for handling long audio inputs from users.

What streaming methods are available?

There are two ways you can stream your transcription depending on your use case and whether you are trying to transcribe an already completed audio recording or handle an ongoing stream of audio and use OpenAI for turn detection:

<a href="https://platform.openai.com/docs/guides/speech-to-text#streaming-the-transcription-of-a-completed-audio-recording" rel="nofollow noopener noreferrer" target="_blank">Streaming the transcription of a completed audio recording</a>

<a href="https://platform.openai.com/docs/guides/speech-to-text#streaming-the-transcription-of-an-ongoing-audio-recording" rel="nofollow noopener noreferrer" target="_blank">Streaming the transcription of an ongoing audio recording</a>

- <a href="https://platform.openai.com/docs/guides/speech-to-text#streaming-the-transcription-of-a-completed-audio-recording" rel="nofollow noopener noreferrer" target="_blank">Streaming the transcription of a completed audio recording</a>
- <a href="https://platform.openai.com/docs/guides/speech-to-text#streaming-the-transcription-of-an-ongoing-audio-recording" rel="nofollow noopener noreferrer" target="_blank">Streaming the transcription of an ongoing audio recording</a>

Note that streaming is not supported with the <code>whisper-1</code> model.

The supported file formats are included <a href="https://platform.openai.com/docs/guides/speech-to-text#overview" rel="nofollow noopener noreferrer" target="_blank">in our API docs</a>.

Can I send links to audio files to the Audio API?

No, you must send a file in one of the supported audio formats.

General questions about the Whisper, speech to text, Audio API

Audio API FAQ

ChatGPT

DALL·E

Service Status

Find answers and get help from Intercom Support and Community Experts

Link, Press control-option-right-arrow to exit

Empty Help Center

Uh oh. That page doesn’t exist.

Disappointed

Neutral

Smiley

Thinking...

Searching through sources...

Analyzing...

Tickets submitted through the messenger or by a support agent in your conversation will appear here.