mehnoorsiddiqui/whatsapp-voice-transcriber

WhatsApp voice transcriber is an audio message transcriber app created with APIMatic auto-generated SDKs for WhatsApp and OpenAI Whisper API. 

WhatsApp Voice Magic

How it works

WhatsApp voice magic is an audio message transcriber app created with APIMatic auto-generated SDKs for WhatsAppand OpenAI Whisper API.This voice transcriber bot can transcribe incoming voice messages in WhatsApp to text in real-time, enabling easy communication across languages.

Running Locally

Cloning the repository to the local machine.

git clone https://github.com/mehnoorsiddiqui/whatsapp-voice-magic

Creating a account on Meta to get an API key.

1. Go to Meta to make a developer account.
2. Create an App by following the getting started guide.
3. You can find your Access token from the Meta for Developers site App Dashboard > WhatsApp > Getting Started.

Setting up Webhook URL

To set up webhook on the Meta Developer portal (also enter matching verify token, app in our case). You may use a service like ngrok to make your local app publicly accessible so this webhook can be registered with Whatsapp. []

Storing the API keys in .env

Replace the WHATSAPP_TOKEN with your token in the .env file. Replace the PHONE_NUMBER_ID with yours in the .env file.

Installing the dependencies.

1. Navigate to the OpenAI+Whisper+API directory and run

npm install

2. Navigate to the WhatsApp+Cloud+API directory and run

npm install

3. Navigate to the root folder and run

npm install

Running the application.

Then, run the application in the command line and it can be tested by sending an audio message to test number given in the developer account.

npm start

CodeSandbox

1. Fork this codesandbox repo.
2. Replace the access token with yours in the .env file.
3. Add the URL of the running sandbox as the webhook URL in the developer account