VidVoice

VidVoice is an AI-powered Streamlit application that generates YouTube scripts and voiceovers based on video titles. It helps content creators automate video production with minimal effort.

Live Demo

VidVoice Streamlit App

Features

• AI-Powered Script Generation: Generates high-quality YouTube scripts based on the video title.
  
• AI Voiceovers: Converts scripts into natural-sounding voiceovers using TTS models.
  
• Multiple AI Models: Supports Google Gemini, Groq, and ElevenLabs APIs for text and speech generation.
  
• Editable Scripts: Modify generated scripts before converting them to voiceovers.
  
• Easy Download: Listen to and download generated voiceovers.
  

Requirements

• Google Gemini API Key
  
• Groq API Key
  
• ElevenLabs API Key (for advanced TTS)
  

Installation

1. Clone the repository:
   bash git clone https://github.com/your-username/VidVoice.git cd VidVoice
2. Install dependencies:
   bash pip install -r requirements.txt
3. Run the Streamlit app:
   bash streamlit run app.py

Usage

1. Generate YouTube Scripts

• Enter the video title.
  
• Set the desired video length.
  
• Click "Generate Script" to create the script.
  
• Edit the script if needed.
  

2. Generate Voiceovers

• Select a TTS model from the sidebar.
  
• Click "Generate Audio" to convert the script into a voiceover.
  
• Download the generated audio.
  

3. Regenerate & Edit

• Click "Regenerate Script" for a new version.
  
• Edit the script and generate audio again if needed.
  

Future Improvements

✅ Local AI Models – Reduce API dependency by integrating local text and voice generation models.
✅ Voice Cloning – Allow users to choose from custom and cloned voices.
✅ Multi-Language Support – Expand voice generation capabilities to multiple languages.

Inspiration

Creating YouTube content can be time-consuming, requiring both scripting and voiceovers. We wanted to simplify this process using AI, making video creation effortless for everyone, from beginners to professionals.

What it does

VidVoice generates high-quality YouTube scripts based on video titles and converts them into natural-sounding voiceovers using AI-powered TTS models. It streamlines content creation by automating scriptwriting and narration in a few clicks.

How we built it

• Frontend: Streamlit for an interactive and user-friendly interface.
  
• AI Models: Google Gemini and Groq for script generation, ElevenLabs for TTS.
  
• Backend: Python for API integration and text processing.
  
• Hosting: Deployed using Streamlit Cloud for easy access.
  

Challenges we ran into

• Ensuring high-quality script generation across different video topics.
  
• Optimizing voiceover output for natural and engaging narration.
  
• Managing API rate limits and response times for seamless performance.
  

Accomplishments that we're proud of

• Successfully automating the entire YouTube content creation workflow.
  
• Integrating multiple AI models for both text and speech generation.
  
• Providing a free and accessible tool for content creators.
  

What we learned

• How to optimize AI-generated content for better readability and engagement.
  
• The importance of voiceover quality in enhancing video production.
  
• Balancing API dependencies while maintaining performance and affordability.
  

What's next for VidVoice

✅ Local AI models to reduce dependency on external APIs.
✅ Voice cloning for custom and personalized narrations.
✅ Multi-language support for a global reach.
✅ Background music & sound effects to enhance audio production.

VidVoice

Generating Voice over audio for Youtube Videos based on Youtube Title

This project is a Streamlit application that leverages various AI models to generate YouTube scripts and voiceovers. It aims to assist content creators in producing high-quality YouTube videos with minimal effort.

• No Elven labs feautre in this Link

Audio Demo

• Title : what is phenomenon beyond northern lights --- Demo Samlpe Link

WorkFlow

Requirements

• Google Gemini Api Key
• Groq API
• ElevenLabs API

Installation

1. Clone the repository:

   git clone https://github.com/your-username/AI-YouTube-Voice-Over-Generator.git
   cd AI-YouTube-Voice-Over-Generator
   

2. Install the required Python packages:

   pip install -r requirements.txt
   

3. Run the Streamlit application:

   streamlit run app.py
   

Usage

API Keys

• Gemini Models: Enter your Gemini API key in the sidebar if you are using Gemini-based text generation models.
• Groq Models: Enter your Groq API key in the sidebar if you are using Groq-based text generation models.
• ElevenLabs TTS: Enter your ElevenLabs API key in the sidebar if you select the ElevenLabs text-to-speech model.

Generating Scripts

1. Enter Video Title: Input the title for your YouTube video.
2. Set Video Length: Specify the desired video length in minutes.
3. Generate Script: Click the "Generate Script" button.
4. Edit Script: Modify the generated script if needed.

Generating Voiceovers

1. Select TTS Model: Choose a TTS model in the sidebar.
2. Generate Audio: Click the "Generate Audio" button after the script is ready.
3. Download Audio: Listen to and download the generated audio.

Regenerating and Editing

• Regenerate Script: Click to create a new script if desired.
• Edit Script: Update the script and convert it to audio.

Contributing

Contributions are welcome! Please open an issue or submit a pull request if you have suggestions for improvements or new features.

Acknowledgments

• mrfakename for Hosting freely Hosting MELLO TTS model, without there contribution it woudn't be a possibilty to provide free TTS service to end users

Improvements

• Local Text Generation: Implement the use of local text generation models to enhance performance and reduce dependency on external APIs.
• Local TTS Models: Integrate local TTS models for better audio generation and faster processing times.
• Voice Cloning: Allow users to choose from a variety of voices for TTS, including options for voice cloning based on user preferences.

Built With

• ai
• elevenlabs-api
• google
• googleai