SpeechRecognitionįor STT (speech-to-text) we are using SpeechRecognition. Run pip install google-api-python-client and follow the instructions on the previous link how to authenticate.
Install Google API Client Library for Python, read more. This is needed, if Google is used in STT and/or TTS. See the Debian/Ubuntu section on the PyAudio installation guide.
Given, that we have Homebrew installed, run brew install portaudio. Install PyAudio (and the required portaudio). We are going to work with Python 3.6 and within a virtualenv, so create one.īecause the installation process is a bit tricky, this time we are going to install the Python requirements one by one, instead of pip install -r requirements.txt. To capture the speech and change that to text, we are going to need a couple of things. The codeĪ simple demo app to test the tech is in demo-app.py. Requires Python 3.6, tested on macOS Sierra and Raspberry Pi 3. My lightweight J.A.R.V.I.S desktop experiment.īuilt mainly for learning purposes, how to implement lightweight speech-to-text and text-to-speech to a Python app.
