How to suppress microphone during voice playback using pocketsphinx/festival

[Ross Munro]

I'm playing with a simple voice assistant model using pocketsphinx and festival. Livespeech listens through the mic and then festival.sayText() plays the phrase back through the speakers. This sets up a charming but useless feedback loop whereby pocketsphinx repeatedly attempts to decipher it's own utterances being picked up through the mic (I assume). I need to suppress the microphone before the festival.sayText() function plays, then restore it after. How do you do that.

import os
from pocketsphinx import LiveSpeech
import festival
import speech_recognition as sr

model_path = '/data/python/Eliza'
print(model_path)

speech = LiveSpeech(
    verbose=False,
    sampling_rate=16000,
    buffer_size=2048,
    no_search=False,
    full_utt=False,
    hmm=os.path.join(model_path, 'en-us-mm'),
    lm=os.path.join(model_path, 'en-us.lm.bin'),
    dic=os.path.join(model_path, 'cmudict-en-us.dict')
)

for phrase in speech:
    festival.sayText(str(phrase))

[Nickolay V. Shmyrev]

You can call speech.ad.stop_recording before calling Festival:

https://github.com/bambocher/pocketsphinx-python/blob/769492da47e41b71e3dd57a6b033fbba79e57032/swig/sphinxbase/ad_base.i#L77