Fact · Speech recognition
60seconds
Azure's REST API for short audio accepts no more than 60 seconds of audio in a request when the audio is sent directly.
Checked against its source on October 17, 2026
In the source's own words
“Requests that use the REST API for short audio and transmit audio directly can contain no more than 60 seconds of audio”
Speech to text REST API for short audio, Microsoft Learn (Azure AI Speech), n.d., Limitations.
Open the original- What it measures
- Short audio is the simple request. For longer files, Microsoft points to the batch service, which is a different API.
- What it does not show
- The limit is for direct transmission. Chunked streaming uses the same endpoint, but check the page for the current rules.