Plugin suggestion to enhance voice recognition

[Rouquier Philippe]

Hello,

I'm developping a plugin that uses VOSK-api to perform voice
recognition. So far the pipeline that includes the above plugin looks
like that :
"pulsesrc ! " \
"queue2 max-size-bytes=4294967294 name=Buffer max-size-time=0 max-size-
buffers=0 ! " \
"audioconvert ! " \
"audiorate ! " \
"audioresample ! " \
"audio/x-raw,format=S16LE,rate=44100,channels=1 ! " \
"vosk name=VoskMain ! " \
"fakesink"

I was wondering if there were plugins that you would suggest to add to
the pipeline that could help enhance the audio (filter noise ? ...) and
ease the work of vosk plugin or make it potentially more accurate.

Thanks for any suggestion.

Philippe