Voice Interface Demo (Python)

Demonstrates voice interface capabilities. Uses the "SpeechRecognition" Python package to convert speech to text.

Prerequisites

The "SpeechRecognition" package depends on the "pyaudio" Python package, which in turn depends on the lower-level "portaudio" library. To install "portaudio":

• On a Mac, use homebrew (brew install portaudio).
• On Windows, use pipwin within an active virtual environment (see installation steps below).

Installation

Fork this repo under your own control, then clone your copy onto your local computer and navigate there from the command-line:

cd voice-interface-demo-py/

Setup

Create and activate a virtual environment:

conda create -n voice-env python=3.8
conda activate voice-env

Windows only:

pip install pipwin
pipwin install pyaudio

Install package dependencies:

pip install -r requirements.txt

Usage

Recognize speech and respond with print statements:

python app/demo.py