Unitra Pi-tervox v1.2 via Google Assistant GRPC

This repository a forked and modified code of the reference sample for the google-assistant-grpc Python package.

This runs as a service on a Raspberry Pi Zero W. Raspi sits nicely with an old Unitra Intervox box. Now it needs a fresh set of paint. Some logo redesign. And a switch between Google Assistant and Alexa.

Script implements the following features:

• Triggering a conversation on pull up on a GPIO
• Audio recording of user queries (single or multiple consecutive queries)
• Playback of the Assistant response
• Conversation state management

Release log

1.0

• Refactor
• Document setup steps
• Trigger conversation on GPIO pull up
• Breathing led class coughed up

1.2

• Boot sound!
• Redo triggering conversation
• Refactor the pushable button
• Param option for boot sound
• Param for no boot sound

1.2.1

• GPIO pin for recroding button now can be supplied via command line option
• Button triggering recording can now be configured as Normally Open Or Closed
• --verbose/-v flag now controls GPIO.setwarnings() in PushButton class
• Update readme with steps to setup service

1.3

• Add sound for start conversation
• Add sound for finished conversation
• Add command line option for start conversation sound file
• Add command line option for end conversation sound file

Road Map

General

• Simple script installing it from scratch

1.3.1

• Add default sound directories

1.3.3

• Add default sounds array

1.3.3

• Add sound dirs command line option

1.4

• Add breathing LEDs indicating the device is listening
• Add booting LEDs animation

1.5

• Add sound for nothing played back

1.6

• Add LEDs option for IFTTT functionality or custom commands

1.7

• Try adding Alexa or Snowboy

Prerequisites

• Python (>= 3.4 recommended)
• A Google API Console Project
• A Google account

Setup

• Install Python 3

  • Ubuntu/Debian GNU/Linux:

    sudo apt-get update
    sudo apt-get install python3 python3-venv
    

  • MacOSX, Windows, Other

• Create a new virtual environment (recommended):

  python3 -m venv env
  env/bin/python -m pip install --upgrade pip setuptools
  source env/bin/activate
  

• Hardware : - Hookup a USB soundcourd via USB OTG cable. - Configure asound to use that card via file ~/.asoundrc

      pcm.!default {
        type asym
        capture.pcm "mic"
        playback.pcm "speaker"
      }
      pcm.mic {
        type plug
        slave {
          pcm "hw:1,0"
        }
      }
      pcm.speaker {
        type plug
        slave {
          pcm "hw:1,0"
        }
      }
  

Service setup

- ssh into the Pi (check wpa_injector script to setup an Image before hand)
- Create pi-terkom directory    
- Cd into it    
- Clone this repo
- copy `run_piterkom.sh` to /opt
- give the script run permissions
- copy piterkom.service to /etc/systemd/system
- you may want to reload the systemd daemon
- enable service via sudo systemctl enable piterkom.service
- start up the service now reboot `reboot now & exit`

Snippet for service setup

mkdir pi-terkom
git clone git@github.com:bhm/Unitra-Pi-tervox.git
cd Unitra-Pi-tervox
cp run_piterkom.sh /opt
chmod ug+x /opt/run_piterkom.sh
cp piterkom.service /etc/systemd/system
systemctl daemon-reload
systemctl enable piterkom.service
systemctl start piterkom.service

Authorization

• Follow the steps to configure the project and the Google account.

• Download the client_secret_XXXXX.json file from the Google API Console Project credentials section and generate credentials using google-oauth-tool:

  pip install --upgrade google-auth-oauthlib[tool]
  google-oauthlib-tool --client-secrets path/to/client_secret_XXXXX.json --scope https://www.googleapis.com/auth/assistant-sdk-prototype --save --headless
  

Run the script

• Install dependencies:

  sudo apt-get install portaudio19-dev libffi-dev libssl-dev
  pip install --upgrade -r requirements.txt
  

• Verify audio setup:

  # Record a 5 sec sample and play it back
  python -m audio_helpers
  

• Run the push to talk sample. The sample records a voice query after a key press and plays back the Google Assistant's answer:

  python3 ./push_red_to_talk
  

Troubleshooting

• Verify ALSA setup:

  # Play a test sound
  speaker-test -t wav
  
  # Record and play back some audio using ALSA command-line tools
  arecord --format=S16_LE --duration=5 --rate=16k --file-type=raw out.raw
  aplay --format=S16_LE --rate=16k --file-type=raw out.raw
  

• Run the sample with verbose logging enabled:

  python -m pushtotalk --verbose
  

• If Assistant audio is choppy, try adjusting the sound device's block size:

  # If using a USB speaker or dedicated soundcard, set block size to "0"
  # to automatically adjust the buffer size
  python -m audio_helpers --audio-block-size=0
  
  # If using the line-out 3.5mm audio jack on the device, set block size
  # to a value larger than the `ConverseResponse` audio payload size
  python -m audio_helpers --audio-block-size=3200
  
  # Run the Assistant sample using the best block size value found above
  python -m pushtotalk --audio-block-size=value
  

• If Assistant audio is truncated, try adjusting the sound device's flush size:

  # Set flush size to a value larger than the audio block size. You can
  # run the sample using the --audio-flush-size flag as well.
  python -m audio_helpers --audio-block-size=3200 --audio-flush-size=6400
  

License

Copyright (C) 2017 Google Inc.

Licensed to the Apache Software Foundation (ASF) under one or more contributor license agreements. See the NOTICE file distributed with this work for additional information regarding copyright ownership. The ASF licenses this file to you under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.