I have a USB mike, called a "Buddy 6G". It shows up under Sound --> Input. Under "Device settings". The "Volume" slider spot is always at the far left ("Softer") and can't be moved. The "Input level" bar never shows anything (it contains no slider spot, and I'm not clear whether this is meant to show the volume of sound coming into the mike).
Yesterday, however, I made some brilliant progress with the microphone and with one of my crucial "transition to Linux" goals: in W10 I use DNS (Dragon Naturally Speaking) dictation software all the time. I followed a lot of the configuration advice at this page, and eventually managed to get things so that I could save a bit of microphone input into a .wav file. None of this configuration stuff (alsamixer) changed the appearance of things under Sound --> Input. NB the command which enabled a 10-second .wav file to be created was:
Code: Select all
arecord -f cd -c 1 -D hw:2,0 -d 10 test2.wav
Anyway, the thing is that my microphone still does not appear to be getting proper data input to the "normal" apps: for example, I tried the online speech engine offered by Google (installing the "Online speech recognition" extension in Chrome). The microphone icon now appears in the right-hand part of the Chrome search box. When I click it it says "Speak now" ... I speak... it says "Listening..." and then "Please check your volume levels".
Also, at the above dictation link there are meant to be two ways of using the DeepSpeech module: one by analysing a sound file (success, as I said above), the other by using microphone input directly. The latter command, namely:
Code: Select all
gst-launch-1.0 -m pulsesrc ! audioconvert ! audiorate ! audioresample ! deepspeech silence-threshold=0.3 silence-length=20 ! fakesink
Can anyone suggest what's going wrong with my microphone configuration?