New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Usage examples for mod_bert #1

Open
gps6304 opened this Issue Aug 18, 2014 · 3 comments

Comments

Projects
None yet
2 participants
@gps6304
Copy link

gps6304 commented Aug 18, 2014

Hi,
Do you have any usage examples for mod_bert? We're trying it out, but are having an issue where our results always end up the same, and we suspect we're not using it correctly. Any demo dialplan out there?

-Gabe

@moises-silva

This comment has been minimized.

Copy link
Owner

moises-silva commented Aug 26, 2014

Hi Gabe,

Sorry for not replying before. I sort of assumed github would email me for this kind of stuff but I just noticed the notification.

You are supposed to call the bert_test application on both ends of the call, it won't work otherwise.

Most of the variables you see below are optional.

bert_timeout_ms is number of milliseconds before timing out trying to find the bit pattern that the application transmits (milliwatt)

bert_window_ms is number of milliseconds of the audio analysis window (each x milliseconds verifies how many errors have occurred, etc)

bert_max_err is the max number of allowed errors once the audio stream is in "sync"

bert_timer=soft tells the application to use FreeSWITCH core soft timer instead of timing transmission based on reception of packets (this should probably be default)

bert_hangup_on_error will hangup the call when audio errors occur

bert_debug_io_file is for debugging the rx/tx audio

Note that this application is pretty naive, no real audio analysis is done beyond verifying the milliwatt PCMU sequence is received on each end (with the configured error tolerance). This means any kind of gain control, echo cancellation, etc may break the results

This is an example I used recently:

<extension name="bert_test">
  <condition field="destination_number" expression="^bert_test$">
    <action application="set" data="bert_timeout_ms=5000"/>
    <action application="set" data="bert_window_ms=1000"/>
    <action application="set" data="bert_max_err=10"/>
    <action application="set" data="bert_timer=soft"/>
    <!--<action application="set" data="bert_hangup_on_error=yes"/>-->
    <!--<action application="set" data="bert_debug_io_file=/tmp/bert_debug_${uuid}"/>-->
    <action application="set" data="jitterbuffer_msec=100:200:40"/>
    <action application="answer"/>
    <action application="bert_test" />
  </condition>
</extension>
@gps6304

This comment has been minimized.

Copy link

gps6304 commented Aug 27, 2014

Hi,
Thanks so much for your response — this makes a lot of sense…still playing with this a bit more, but very excited about the potential!

-Gabe

On Aug 26, 2014, at 12:35 AM, Moises Silva notifications@github.com wrote:

Hi Gabe,

Sorry for not replying before. I sort of assumed github would email me for this kind of stuff but I just noticed the notification.

You are supposed to call the bert_test application on both ends of the call, it won't work otherwise.

Most of the variables you see below are optional.

bert_timeout_ms is number of milliseconds before timing out trying to find the bit pattern that the application transmits (milliwatt)

bert_window_ms is number of milliseconds of the audio analysis window (each x milliseconds verifies how many errors have occurred, etc)

bert_max_err is the max number of allowed errors once the audio stream is in "sync"

bert_timer=soft tells the application to use FreeSWITCH core soft timer instead of timing transmission based on reception of packets (this should probably be default)

bert_hangup_on_error will hangup the call when audio errors occur

bert_debug_io_file is for debugging the rx/tx audio

Note that this application is pretty naive, no real audio analysis is done beyond verifying the milliwatt PCMU sequence is received on each end (with the configured error tolerance). This means any kind of gain control, echo cancellation, etc may break the results

This is an example I used recently:


Reply to this email directly or view it on GitHub.

@moises-silva

This comment has been minimized.

Copy link
Owner

moises-silva commented Nov 27, 2014

Just for the record. You can also call bert on one end of the call and echo or delay_echo on the other end

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment