Azure Speech Service Integration in Angular - Testing App

This repository contains a small Angular application that demonstrates the integration of Azure Cognitive Services for speech recognition and speech synthesis. The application uses Azure's voice detection AI to interact with users via text-to-speech and speech-to-text functionalities. This project is intended for testing and demonstration purposes.

See post in Porfolio

Features

• Text-to-Speech: Converts text prompts to speech using Azure's Speech Service and plays the audio output.
• Speech-to-Text: Listens to the user's speech through the microphone and converts it to text, filling in the specified form fields.
• Form Autofill: Automatically fills in form fields such as first name, last name, and age based on user speech input.

Getting Started

Prerequisites

• Node.js (version 12 or higher)
• Angular CLI (version 12 or higher)
• An Azure account with a Speech Service resource

Installation

1. Clone the repository:

   git clone https://github.com/YevaGalstyan/speech-detection.git
   cd speech-detection

2. Install dependencies:

   npm install

3. Set up Azure Speech Service:

   • Sign in to the Azure Portal.
   • Create a new Speech Service resource.
   • Note down the subscription key and service region.

4. Configure the application:

   • Open src/app/speech.service.ts.
   • Replace "6d345534e918442f8dc7a8e1c0df8b53" with your Azure Speech Service subscription key.
   • Replace "westeurope" with your Azure service region.

   const speechConfig = SpeechConfig.fromSubscription("your_subscription_key", "your_service_region");

5. Run the application:

   ng serve

   • Navigate to http://localhost:4200/ in your web browser.

Usage

1. Enter the form control names for first name, last name, and age.
2. The application will use text-to-speech to prompt you to say the values for these fields.
3. Speak the required information when prompted.
4. The application will use speech-to-text to recognize your speech and fill in the form fields automatically.

License

This project is licensed under the Apache License, Version 2.0. See the LICENSE file for more details.