Testing Google API for speech recognition

Requirements

In case you are working in a virtual environment:

pip install SpeechRecognition # importing speech recognition package from google api 
pip install playsound # to play saved mp3 file
pip install gTTS # google text to speech 
pip install requests # to get API responses

Now, the folowing can only be installed as an administrator:

sudo apt install python-gobject # to play saved mp3 file
sudo apt install python3-pyaudio # to play saved mp3 file
sudo apt-get install portaudio19-dev python-pyaudio

Now you can install the following:

pip install PyAudio

As we are working with our own virtual environment we must create a link to some libraries of those packages we just installed as sudo into our environment.

# Link to gi library (from python-object)
ln -s /usr/lib/python3/dist-packages/gi/ /home/iggy/.pyenv/versions/ds/lib/python3.6/site-packages/

Install Flask:

pip install Flask

Tested and working on Ubuntu 18.0.4

Usage

git clone https://github.com/iggyrrieta/gtt_speech_recognition

Go to scripts folder and run from terminal:

python app.py

Go to http://127.0.0.1:5000/