Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Vietnamese TTS with instant voice cloning • On-device • Real-time CPU inference • 24kHz audio quality • Chuyển văn bản thành giọng nói tiếng Việt • Text to speech tiếng Việt • TTS tiếng Việt
| Date | Stars |
|---|---|
| 2026-07-24 | 2238 |
| 2026-07-25 | 2243 |
| 2026-07-28 | 2243 |
| 2026-07-30 | 2243 |
| 2026-07-31 | 2266 |
| 2026-08-06 | 2266 |
Today
— stars today
This week
+23 stars this week
This month
— stars this month
Momentum
23.0
growth rate 1.03%/day
# 🦜 VieNeu-TTS [](https://github.com/keon/awesome-nlp) [](https://discord.gg/yJt8kzjzWZ) [](https://colab.research.google.com/drive/1b9PO-lcGZX9pEkEwQmu8MfhSnjxKrALW?usp=sharing) [](https://huggingface.co/pnnbao-ump/VieNeu-TTS-v3-Turbo) [](https://huggingface.co/pnnbao-ump/VieNeu-TTS-v2) [](https://huggingface.co/pnnbao-ump/VieNeu-TTS) <img width="1087" height="710" alt="image" src="https://github.com/user-attachments/assets/5534b5db-f30b-4d27-8a35-80f1cf6e5d4d" /> **VieNeu-TTS-v2** is the next generation of on-device Vietnamese TTS, featuring **10,000+ hours** of bilingual training, **instant voice cloning**, and a dedicated **Podcast/Conversation** mode. > [!NOTE] > **🆕 VieNeu-TTS v3 Turbo (early access) is out for preview!** > A brand-new architecture **designed and trained from scratch by Phạm Nguyễn Ngọc Bảo** (codec: [MOSS-Audio-Tokenizer-Nano](https://huggingface.co/OpenMOSS-Team/MOSS-Audio-Tokenizer-Nano); phonemizer: [sea-g2p](https://github.com/pnnbao97/sea-g2p)): > - **48 kHz** high-fidelity audio (up from 24 kHz). > - **Built-in default voices** — stable and consistent, no reference clip needed. > - **Reading styles**: natural, news, and storytelling. > - **Emotion / non-verbal cues** *(experimental)*: drop `[cười]`, `[thở dài]`, `[hắng giọng]` straight into the text. > - **Batched generation** (batch size up to 32), including a multi-speaker **Conversation** mode that batches the whole script regardless of speaker. > - **Instant voice cloning** from a 3–8s clip, with automatic reference denoising. > > Try it in the Web UI (backbone **"VieNeu-TTS-v3-Turbo (Thử nghiệm)"**) or the SDK (`Vieneu(mode="v3turbo")`). The **full v3** release is coming in the next few weeks. > [!IMPORTANT] > **🚀 VieNeu-TTS-v2 is here!** > The full high-fidelity bilingual architecture is now available with: > - **10,000+ Hours of Data:** Unmatched naturalness in both English and Vietnamese. > - **Podcast & Dialogue Mode:** Multi-speaker support with emotional nuances. > - **Zero-shot Cloning:** Clone any voice in 3-5 seconds across all v2 variants. ## ✨ Key Features - **10,000+ Hours Training**: Trained on a massive English-Vietnamese dataset for human-like prosody. - **Bilingual (En-Vi) Code-switching**: Powered by [**sea-g2p**](https://github.com/pnnbao97/sea-g2p) for high-fidelity pronunciation and seamless transitions between Vietnamese and English. - **Podcast & Conversation Mode**: Multi-speaker dialogue support with automatic character detection. - **Instant Voice Cloning**: Clone any voice with just **3-5 seconds** of reference audio. - **Ultra-Fast Performance**: Optimized for **GPU (LMDeploy)** and **CPU (GGUF/ONNX)**. - **Production-Ready**: High-quality 24 kHz waveform generation, fully offline. [<img width="600" height="595" alt="VieNeu-TTS Demo" src="https://github.com/user-attachments/assets/021f6671-2d7f-4635-91fb-88b2ab0ddbcd" />](https://github.com/user-attachments/assets/021f6671-2d7f-4635-91fb-88b2ab0ddbcd) ## 📌 Table of Contents 1. [🦜 Installation & Web UI](#installation) 2. [📦 Using the Python SDK](#sdk) 3. [🐳 High-Quality Server (Standard Mode)](#docker-remote) 4. [🔬 Model Overview](#backbones) 5. [🚀 Roadmap](#roadmap) 6. [🤝 Support & Contact](#support) 7. [📑 Citation](#citation) --- ## 🦜 1. Installation & Web UI <a name="installation"></a> ### Setup with `uv` (Recommended) `uv` is the fastest way to manage dependencies. ```bash # Windows: powershell -c "irm https://astral.sh
Excerpt of 21,024 characters
Read on GitHub
310
228
30
11
10
8
5
4
2
2
1
Phạm Phú Ngọc Trai
1
1
1
1
1
Sense_wang
1
Would you bet a product on this? Bounded 0–100 and slow moving.
matched fp:ccf424cedcb1d3d4, topic:text-to-speech, topic:tts, topic:speech-synthesis