🚨 NZG73 strictly instructs all users to use this software only for positive and constructive purposes.
Misusing someone's voice, targeting individuals, or causing harm to anyone's life is highly unethical and a grave offense.
We must fear the consequences of such actions and avoid them. This tool is designed to simplify complex tasks; please use it responsibly. 🤝
🌟 Key Features
The system is divided into several advanced modules to ensure optimal performance for every task. Detailed descriptions of each feature are provided below:
🔊 Step 1: TTS CPU (Text-to-Speech)
This module is specifically designed for users who do not have expensive Graphics Cards (GPUs).
🏷️ Feature
📝 Description
🌍 Supported Languages
Primarily supports 2 languages (English and Chinese)
⚡ Performance
Runs smoothly on a CPU without any lag. No heavy GPU needed.
🎙️ Voice Cloning
Includes a highly functional Voice Cloning feature (Strictly for ethical use only)
🎤 Step 2: Voice Clone (Advanced)
An extremely powerful Voice Cloning module supporting a vast range of languages.
🏷️ Feature
📝 Description
🌍 Supported Languages
Supports 600 different languages 🌐
💻 Hardware Requirements
Due to its heavy and advanced nature, it is very slow on a CPU
🚀 Performance
A GPU is mandatory for best results. Runs easily on cards with 4GB to 8GB VRAM. The more powerful the GPU, the faster the processing speed.
💎 Step 3: Voice Clone Metadata (Special Feature)
🌟 This is the most specialized and advanced part of the software!
Previously, users had to record audio and create text files separately for Voice Training—a process that took hours.
✨ "Dataset preparation is now easier than ever!" ✨
Instead of manually cutting hundreds of clips and naming them, this Automated System handles everything. You simply:
✍️ Enter your text
🖱️ Click a button
🎧 The system generates the voice
📁 Saves it in a numbered sequence in a folder
📝 Automatically updates your metadata file
🎯 Training a dataset is now child's play! 🧒
📝 Step 4: Speech to Text (ASR)
This module converts audio into text with incredible accuracy.
🏷️ Feature
📝 Description
🌍 Supported Languages
Supports 99 languages
💻 Performance & Hardware
This is a powerful model requiring at least a 3GB VRAM GPU. While it works excellently on a GPU, it is nearly impossible to use on a CPU due to extremely slow speeds.
🔄 Step 5: Voice to Voice
This feature converts one voice directly into another.
🔧 How it works:
Suppose you generated a story or audio using an AI model (Target Voice), but you want it to sound like your own voice. You provide:
🎯 The Target Voice (AI generated audio)
🎙️ A 30-second reference clip of your Original Voice
After clicking Generate, the entire audio is transformed into your voice! 🎉
⚡ Performance: Somewhat slow on a CPU but functional. However, on a GPU, it processes with fluid speed. 🚀
🚀 Upcoming Releases
The journey of NZG73 doesn't end here. We are bringing you more exciting updates:
📱 N Preva (Mobile App)
Our mobile app, N Preva, will be released soon, bringing powerful AI features directly to your smartphone.
🌐 Advanced Web UI
A new, advanced Web UI is in development. It will support multiple AI models, allowing you to interact with various AI systems and handle different tasks simultaneously.
🛠️ Powered By 🛠️
🌟 Connect With Me 🌟
📊 GitHub Stats 📊
✨ Made with ❤️ by Muhammad Noman | NZG73 ✨
⭐ If you like this project, don't forget to give it a star! ⭐
-000000?style=for-the-badge&logo=x&logoColor=white&labelColor=black){kind=icon}
