Parameter Type Default Description
audioChannels int 1 Audio channels for audio source
audioSampleRate int 16000 Audio sample rate of audio source
vadAggressiveness int 1 Noise filtering level. Accepts 1 thru 3.
speechRatio float 0.75 Must be between 0 and 1 as a decimal
speechBufferSize int 50 Buffer size for speech frames
speechBufferPadding int 350 Padding, in milliseconds, of speech frames
audioDeviceIndex int 0 Microphone device accourding to PyAudio
speechModel str None Path and filename of Coqui Speech Model file
speechScorer str None (Optional) Path and filename of Coqui Scorer file
nickname str None (Optional) Nickname of the device

Help & Support

Help and additional details is available at