Hi everyone,
please could you support me with the problem of sound?
There is problem with sound (MediaStream) in the Google Chrome browser turns out to be distorted and illegible when I use the sound of the audio capture device (RemoteAudio) (sound metal or like robot).
It was checked with:
-Simple Audio Recorder
- Facebook voice calls
- Talky.io (WebRTC service).
I’ve used the Google Chrome browser - version 64.0.3282.186 official build, 64 bit on the remote desktop of Windows 10/ Intel Xeon CPU CPU E5-2673 v3 2.40GHz - 4 cores and 8 GB RAM without any hardware sound support. Chrome uses about 10-20% of CPU with WebRTC and I have about 5Gb free memory.
In other applications, this problem does not occur and the captured audio sound with quite high quality.
I checked it with:
-Native applications (Sound Recorder App, Voice Recorder, Skype)
-oCam Screen Recorder 366.0
My assumptions are:
1. In fact the sound comes from the network (remote audio) and some sound frames are formed with a delay. At some moments there are no sound frames. Chrome has small buffer size for micropthone so nothing can be encoded .
2. Without hardware support, Chrome does not have enough time to encode the audio stream (encoding the audio stream).
The increased constant latensy for captured sound, more CPU and memory usage, unstable operation of the browser itself could happen. Perhaps I can do something using Windows, for example set priority of process for a realtime (Chrome runs a lot processes and I just don’t know what kind of a specific speaker can be responsible for the Mediastream) or check input audio stream.
What do you think are there any extensions (filters for capture audio), tweaks or experimental settings (keys or flags) for Chrome that can change the captured audio stream?
What tutorial/ resources could be useful to solve this problem?