I’m excited to share my latest workspace project: Translate and Voice Vietnamese TTS. This tool is designed for real-world applications, focusing on high-quality Vietnamese speech synthesis and automated video localization.
Key Features:
Vietnamese Text-to-Speech (TTS): High-quality voice generation with multiple runtime configurations.
Voice Cloning: Create a digital voice profile using just a short reference audio sample.
End-to-End Video Dubbing: Automatic audio extraction, speech recognition, translation into Vietnamese, and re-dubbing.
Real-time Streaming: Supports low-latency audio generation for fast-response scenarios.
Subtitle Burn-in: Option to hardcode Vietnamese subtitles directly onto the output video.
Smart Runtime: Includes model caching and auto-restore to quickly resume your last session.
Technical Foundation: Built upon the excellent open-source work of VieNeu-TTS by pnnbao97.
Hardware Requirements:
NVIDIA GPU: CUDA version >= 12.0
Apple Silicon: MPS + 16 GB RAM (Minimum)
🔗 Explore the Project on GitHub: https://github.com/levinhtuyen/Translate-and-Voice-Vietnamese-TTS
Top comments (0)