SoundWave: Speech Assistance For People With Stuttering

Stuttering is a speech disorder that disrupts fluency, causing pauses, repetitions, and difficulty in verbal communication. Individuals with stuttering often face challenges in social, educational, and professional settings due to speech interruptions. Existing speech assistance tools lack real-time processing and personalization, limiting their effectiveness in providing immediate support. This project aims to develop SoundWave: A Speech Assistance System for People with Stuttering, which enhances communication fluency through artificial intelligence.

The system integrates Speech-to-Text (STT) using the Vosk ASR model, which converts spoken words into text. Next-Word Prediction, powered by GPT-2, anticipates the user’s intended word based on context, reducing pauses and hesitations. Text-to-Speech (TTS) using YourTTS then converts the predicted text into speech while maintaining natural intonation. Additionally, Voice Cloning technology ensures the generated speech retains the user’s original voice, making interactions more personal and seamless. SoundWave offers an innovative, assistive solution that empowers individuals with stuttering, enhances their communication skills, and improves their quality of life. This system has potential applications in therapy, education, and professional environments, making speech assistance more accessible and effective.

Features

• Speech-to-Text (STT): Converts the user’s spoken words into text using Automatic SpeechRecognition (ASR).
• Next-word Prediction: Predicts the next word the user is likely to say based on context using GPT-2 model
• Text-to-Speech (TTS): Converts the predicted text into natural-sounding speech using YourTTS.
• Voice Cloning: Generates speech using the user’s own voice rather than a robotic or generic voice with minimal training data.
• Real-Time Inference: Optimized for fast and efficient speech synthesis.

Getting Started

Installation

1. Clone the repository:

   git clone https://github.com/sreenaddh/SoundWave.git
   cd SoundWave

2. Install dependencies:

   pip install -r requirements.txt

3. Run the application:

   python server.py  # Running local server
   cd soundwave
   flutter run  # For running app on pc