Raspberry Pi Setup for Voice Assistant Applications Using ChatGPT, Whisper API, gTTS and Pysstx3

 

Raspberry Pi can be used for providing voice assistant capabilities by integrating with ChatGPT and Whisper APIs from OpenAI.

This article shows how to set up the required libraries such as Chat GPT, Whisper API, Text-to-speech Pysttx3 etc., on Raspberry Pi for enabling voice assistant  applications.

 

Hardware components

• Raspberry Pi
• Micro SD Card
• Power Supply
• USB Speaker
• USB Microphone

• Recommended for initial setup
• USB Mouse
• USB Keyboard
• HDMI Cable
• Monitor
• • Software Apps:
    • Python
    • ChatGPT, Whisper, Speech Recognition, Pysttx3 libraries

Raspberry pi operating system needs to be installed on a micro SD card before installing any ChatGPT based libraries. Raspberry Pi Imager running on another computer can be used to copy the operating system into the SD card. 

Click on 'CHOOSE OS' button and select Raspberry Pi OS (64-bit) option and select 'WRITE' button to install the operating system on the SD card.

After installing Raspberry pi os on SD card, it can be inserted into Raspberry pi and connect to Monitor and keyboard for installing the required voice assistant integrations. First setup username and password.

Select Preferences -> Raspberry Pi Configuration.

Click on the Interfaces tab and select SSH and VNC. These options will enable you to access Raspberry pi remotely using SSH or VNC.

Advanced configurations can be set using raspi-config comand on the Terminal.

sudo raspi-config

Select System -> Audio -> USB Audio

Sound settings can be modified using alsamixer command on the Termial.

                           

Install openai library and then install gpt_index which is also known as LlamaIndex (GPT Index) is a project that provides a central interface to connect your LLM's with external data. Install PyPDF2 library which can be used for processing PDF files.

pip install openai==0.27.4
pip install gpt_index==0.4.24
pip install PyPDF2

Install longchain library which is a framework for developing applications powered by language models. Also install tiktokpn, a tokenizer library and PyCryptodome which provides encryption and decryption algorithms.

 pip install langchain==0.0.132
pip install PyCryptodome
pip install tiktoken==0.3.3




sudo apt-get install libasound-dev
sudo apt-get install portaudio19-dev
pip install PyAudio

Install epseak, speechRecoginition and text to speech library (pyttsx3) libraries.

sudo apt-get install espeak
pip install SpeechRecognition
pip install pyttsx3

Install gTTS (Google Text-to-Speech) library.

pip install gTTS
pip install playsound
pip install pygobject