Skip to end of metadata
Go to start of metadata

TALK_DETECT()

Synopsis

Raises notifications when Asterisk detects silence or talking on a channel.

Description

The TALK_DETECT function enables events on the channel it is applied to. These events can be emited over AMI, ARI, and potentially other Asterisk modules that listen for the internal notification.

The function has two parameters that can optionally be passed when set on a channel: dsp_talking_threshold and dsp_silence_threshold.

dsp_talking_threshold is the time in milliseconds of sound above what the dsp has established as base line silence for a user before a user is considered to be talking. By default, the value of silencethreshold from dsp.conf is used. If this value is set too tight events may be falsely triggered by variants in room noise.

Valid values are 1 through 2^31.

dsp_silence_threshold is the time in milliseconds of sound falling within what the dsp has established as baseline silence before a user is considered be silent. If this value is set too low events indicating the user has stopped talking may get falsely sent out when the user briefly pauses during mid sentence.

The best way to approach this option is to set it slightly above the maximum amount of ms of silence a user may generate during natural speech.

By default this value is 2500ms. Valid values are 1 through 2^31.

Example:

same => n,Set(TALK_DETECT(set)=) ; Enable talk detection

same => n,Set(TALK_DETECT(set)=1200) ; Update existing talk detection's silence threshold to 1200 ms

same => n,Set(TALK_DETECT(remove)=) ; Remove talk detection

same => n,Set(TALK_DETECT(set)=,128) ; Enable and set talk threshold to 128

This function will set the following variables:

Note

Icon

The TALK_DETECT function uses an audiohook to inspect the voice media frames on a channel. Other functions, such as JITTERBUFFER, DENOISE, and AGC use a similar mechanism. Audiohooks are processed in the order in which they are placed on the channel. As such, it typically makes sense to place functions that modify the voice media data prior to placing the TALK_DETECT function, as this will yield better results.

Example:

same => n,Set(DENOISE(rx)=on) ; Denoise received audio

same => n,Set(TALK_DETECT(set)=) ; Perform talk detection on the denoised received audio

Syntax

TALK_DETECT(action)
Arguments
  • action
    • remove - W/O. Remove talk detection from the channel.
    • set( dsp_silence_threshold,dsp_talking_threshold ) - W/O. Enable TALK_DETECT and/or configure talk detection parameters. Can be called multiple times to change parameters on a channel with talk detection already enabled.
      • dsp_silence_threshold - The time in milliseconds before which a user is considered silent.
      • dsp_talking_threshold - The time in milliseconds after which a user is considered talking.

See Also

Import Version

This documentation was imported from Asterisk Version GIT-13-8cc34aa

  • No labels

4 Comments

  1. Eyal Hasson

    I thing there is a mistake on the descriptions of the parameters. If I understand correctly the first is the time of silence to be identified as end of talk, and the second is the energy to be considered as talking.

    1. Rusty Newton

      I don't believe that is accurate. I think the descriptions are correct. If you have reason to believe differently - you can file a bug on issues.asterisk.org/JIRA and explain the issue there.

      Thanks!

  2. Eyal Hasson

    From the review board summery: https://reviewboard.asterisk.org/r/3563/

    "The function allows setting both the silence threshold (the length of silence after which we decide no one is talking) as well as the talking threshold (the amount of energy that counts as talking):"

    One of them is mistaken. From my (very short) experience, it seems the review board version is correct.

    1. Rusty Newton

      Feel free to file a bug on the issue tracker (issues.asterisk.org/jira) if you found the documentation to be inaccurate. Thanks!