Hello Ubuntu community,
I am using Ubuntu 26.04 LTS with GNOME. My internal microphone is detected correctly by the system, but I have a problem with browser-based voice applications.
When I use:
-
ChatGPT Voice
-
YouTube voice search
-
Other browser-based speech recognition services
the microphone works, but the audio quality is very poor.
Symptoms:
-
My voice sounds distorted.
-
Speech recognition often misunderstands words.
-
The microphone input seems noisy or incorrectly processed.
-
The problem is mainly noticeable inside the browser.
System information
Ubuntu: 26.04 LTS
Desktop Environment: GNOME
Audio System: PipeWire
PipeWire version: 1.6.2
WirePlumber version: 1.6.2
Microphone hardware
Internal microphone:
ALC256 Analog
Ryzen HD Audio Controller
The microphone is detected by ALSA:
**** CAPTURE Hardware Devices ****
card 1: Generic_1 [HD-Audio Generic]
device 0: ALC256 Analog [ALC256 Analog]
PipeWire status
wpctl status -n shows:
Sources:
* 59. alsa_input.pci-0000_04_00.6.analog-stereo [vol: 1.00]
Firefox is detected by PipeWire:
Firefox [1.6.2]
What I have checked
-
Microphone permissions in the browser are enabled.
-
The microphone is detected by the system.
-
PipeWire and WirePlumber are running correctly.
-
The issue is not that the microphone is missing; the problem is the input quality.
Questions
-
Are there recommended PipeWire or WirePlumber settings for ALC256 microphones on Ubuntu 26.04 LTS?
-
Could microphone boost, sample rate, or audio profile settings cause distorted input in WebRTC applications?
-
Is there any Firefox-specific configuration needed for better PipeWire microphone handling?
-
Should I adjust settings using
pavucontrol,wpctl, or another modern PipeWire tool?
I can provide additional logs or command outputs if needed.
Thank you for your help.
