VoiceBotAI

Voice-Interactive Chatbot

This program integrates IBM Watson's Speech to Text and Text to Speech services with OpenAI's GPT-3.5 model, providing a seamless voice-interactive chatbot experience.

Features

• Speech to Text: Converts voice to text using IBM Watson's API.
• ChatGPT Interaction: Processes the text with OpenAI's GPT-3 to generate responses.
• Text to Speech: Converts the chatbot's text responses back into voice using IBM Watson's API.

Prerequisites

• IBM Cloud account with Watson Speech to Text and Text to Speech services.
• OpenAI account with GPT-3 API access.
• Python 3.x.

Ensure you have the API keys for IBM Watson and OpenAI, and install the required Python packages.

Installation

1. Clone or download this repository.

   git clone git@github.com:maxstocklin/VoiceBotAI.git 
   

2. Edit Bash Profile:

   vim ~/.zshrc 
   

3. Set your API keys as environment variables:

    export STT_API_KEY='your-ibm-speech-to-text-key'
    export TTS_API_KEY='your-ibm-text-to-speech-key'
    export OPENAI_API_KEY='your-openai-key'

4. Install Python packages

   pip install --upgrade openai
   pip install --upgrade ibm-watson
   pip install --upgrade ibmcloud

Usage

Run the script with the path to your audio file:

 python chatbot.py <audiofile_path>

Replace <audiofile_path> with your audio file's path.

How It Works

Step 1: Speech to Text

Transcribes audio to text using IBM's API.

Step 2: ChatGPT Interaction

Sends text to ChatGPT for response.

Step 3: Text to Speech

Converts response to speech using IBM's API.