Reputation: 1
I have a setup where I open a connection to freeswitch through the ESL and start exchanging commands.
In one specific scenario I want for freeswitch to call me and record a message. So I call a phone number with sofia and park the call
originate {set some private variables and origination_caller_id_number}sofia/gateway// &park()
During the call I play a few messages
uuid_broadcast playback::
And listen to events waiting specific for DTMF tones so I can take action. Play another message or start recording
To stop a playback and start recording
uuid_break uuid_record start
I also playback the recorded file to the user using the same playback command
Now the issue, the first time a message is recorded it works fine, I can listen to it. After I record a new message on the same call nothing is recorded in the file. I can download the file to listen to it directly and still no sound. I see that the file is created and it's size is compatible with the length recorded but even looking with Audacity there is no audio in it.
What can be causing this and does anyone have an idea on how to fix it?
Thanks for the help!
Upvotes: 0
Views: 720
Reputation: 1
I have reproduced this issue as well. I believe this is a bug in FreeSwitch. I am not aware of any work around.
If you want to hear audio in the second recording, you must make the recording LONGER than the first recording. I believe the reason for this is that the audio buffer from the first recording is still associated to the recording session and it is full of silence. The stale first audio buffer is saved at the beginning of the second audio recording. So if you want to hear anything from the second recording, make it a longer recording.
Upvotes: 0
Reputation: 1961
this looks like a bug, probably worth submitting to Jira. Do you specify a different file name for a new recording?
Upvotes: 0