Hi,
I am using webrtc c++ librarry in Raspberry Pi Zero 2W to stream audio to Janus audio bridge plugin.
While on live connection, the audio from Pi switches between normal voice to robotic kind of voice automatically, and keep on switching between normal and robotic.
I implemented custom ADM push / pull audio through webrtc. The ADM opens ALSA device and captures audio from mic (in sync with available frames: snd_pcm_avail_update), and pushes through webrtc::AudioTransport#RecordedDataIsAvailable() in ADM. I am also passing in translated optional timestamp as last parameter to RecordedDataIsAvailable().
What could cause robotic like voice as in attached wav file?
I recorded the attached audio file in my local PC. So the path of original audio is: Pi -> ALSA -> WebRTC -> Janus -> WebRTC -> My local PC -> Playback -> Record played audio.
Thanks