Utterance Split

Utterance Split detects pauses between words in submitted audio. Used when the Utterances feature is enabled for pre-recorded audio.

Pre-recorded Streaming

Deepgram’s Utterance Split feature monitors incoming audio and detects when a sufficiently long pause is detected between words. By default, the length of time Deepgram uses for Utterance Split is 0.8 seconds, but you can configure this value using the utt_split parameter.

Utterance Split is used when the Utterances feature is enabled for pre-recorded audio.

Enable Feature

To enable Utterance Split, when you call Deepgram’s API, add an utt_split parameter in the query string and set it to the length of time (in seconds) of silence between words after which Deepgram will decide that a new utterance should begin. The default values is 0.8 s.


To transcribe audio from a file on your computer, run the following curl command in a terminal or your favorite API client and define the utterance split value you'd wish to use:

curl \
  --request POST \
  --header 'Authorization: Token YOUR_DEEPGRAM_API_KEY' \
  --header 'Content-Type: audio/wav' \
  --data-binary @youraudio.wav \
  --url 'https://api.deepgram.com/v1/listen?utterances=true&utt_split=LENGTH-OF-TIME-IN-SECONDS'

:eyes: Replace YOUR_DEEPGRAM_API_KEY with your Deepgram API Key.


To learn about the results, see Utterances.

Use Cases

Some examples of use cases for Utterance Split include:

  • Audio with speakers who pause longer while speaking than the average speaker.
  • Audio with speakers who speak with shorter pauses than the average speaker.