Published March 13, 2020 © CC BY-NC

Monty Python's Argument Bot

In need of a good argument? Try the argument bot! A robot based on Monty Python's Flying Circus argument sketch.

IntermediateFull instructions providedOver 1 day4,605

Things used in this project

Hardware components

Raspberry Pi 3 Model B

Google AIY Kit

Tower Pro MG995

Software apps and online services

Google Dialogflow

Photo Editing Software

Hand tools and fabrication machines

3D Printer (generic)

Scissors, Free Fall

Multitool, Screwdriver

Screw

Glue

Printer

Story

Original Monty Python sketch

Project Video

Head

First things first, we need a person to argue with, and there is none better than the face of the original argument consultant, Mr. John Cleese.

After finding a nice picture online, we use photo editing software to cut out his head and create two separate parts; just his chin, and his full face with a black patch for a mouth.

1 / 2

Once we are happy with the result, we print said parts on some thick paper.Now all that's left to do is cut neatly along the lines and try out our newly created cutout animation!

Electronics

1 / 2

Making Mr. Cleese talk by hand is fun, but it isn't a robot just yet. For that we need some electronics:

Raspberry Pi 3B+
Google AIY Voicekit
Tower Pro MG995

There are great tutorials on getting started with the AIY Kit and working with servos.

After following these guides we are ready to move on the next bit, well done!

Linear Actuator

1 / 8

The crux of this projects is the linear actuator. It is an addition to our servo to make sure we can move the chin up and down. We don't want to reinvent the wheel so we used this one, thanks davelnewton!

We made some minor changes to make it stand on itself and make it easy to attach both the chin and face.

The altered linear actuator is added in attachment.

After assembling and gluing everything we can also run the test code, look at him wag his chin!

Dialogflow

1 / 3

With all the physical work done, now it's time to add the ability to argue. Queue Dialogflow, Google's chatbot service. They also have a great quick start.

After all the setup is done, it's time to add all our intents, the more the merrier!

Overview and Code

Last but not least we need to write some code. No worries, it's added as well.Here is an overview of how the full project works:

1The microphone picks up someone speaking and records the audio.

2-3Using some Google magic (Speech-To-Text) we extract the text from the audio.

4-5This text is sent to our chatbot (Dialogflow) and is matched with an intent. After which one of the possible answers is sent back to the Raspberry Pi.

6Using Text-To-Speech, the text is converted to audio and played via the speaker, while we also move the chin up and down.

Result!

We have a strict looking face with lifelike movements and the ability to counter any position, let the bickering commence!

Schematics

Code

arguCode.py

# Servo setup
from gpiozero import Servo
mouthServo = Servo(26)
global talk

# Generic imports
import time
import os
import threading
from random import randint

# Dialogflow setup
import aiy.voice.tts 
import dialogflow_v2 as dialogflow
from aiy.board import Board
from aiy.cloudspeech import CloudSpeechClient

session_client = dialogflow.SessionsClient()
session = session_client.session_path('xxxxxxx', 12345)

def getDialogResponse(text):

    text_input = dialogflow.types.TextInput(text=text, language_code='EN')
    query_input = dialogflow.types.QueryInput(text=text_input)
    response = session_client.detect_intent(session=session, query_input=query_input)
    text = response.query_result.fulfillment_text

    return response

def talk():

    global talk
    while True:
        while talk:
            sleepTime = (randint(1, 3))/10
            mouthServo.max()
            time.sleep(sleepTime)
            mouthServo.mid()
            time.sleep(sleepTime)
        
        mouthServo.detach()

moveLidThread = threading.Thread(target=talk)
moveLidThread.start()

def say(test):
    aiy.voice.tts.say(test, lang='en-GB', volume=50, pitch=60, speed=80, device='default')

while True:

    print('Listening...')
    spokenText = client.recognize() 

    if spokenText is None:
        print('You said nothing.')
    else:
        response = getDialogResponse(spokenText)
        responseText = response.query_result.fulfillment_text

        talk = True
        say(responseText)
        talk = False

    time.sleep(0.5)

Monty Python's Argument Bot

Things used in this project

Hardware components

Software apps and online services

Hand tools and fabrication machines

Story

Original Monty Python sketch

Project Video

Head

Electronics

Linear Actuator

Dialogflow

Overview and Code

Result!

Custom parts and enclosures

Linear Actuator

Schematics

Dataflow

Code

arguCode.py

Credits

Comments

Embed the widget on your own site

Monty Python's Argument Bot

Monty Python's Argument Bot

Things used in this project

Hardware components

Software apps and online services

Hand tools and fabrication machines

Story

Original Monty Python sketch

Project Video

Head

Electronics

Linear Actuator

Dialogflow

Overview and Code

Result!

Custom parts and enclosures

Linear Actuator

Schematics

Dataflow

Code

arguCode.py

Credits

Comments

Related channels and tags