🗣️ Hmong TTS — Orpheus‑3B version 2 best hmong TTS (Text to Speech) (Fine‑Tuned) | LocalVoice.org
license: apache-2.0
Hmong Text‑To‑Speech (TTS) model fine‑tuned from Orpheus‑3B‑TTS, optimized with Unsloth + SNAC codec. Built by LocalVoice.org to support Hmong language technology.
🙏 Special thanks to ThaiSC & HPC Ignite Program for HPC resources.
🌟 Model Highlights
- ⚙️ Base Model: Orpheus‑3B TTS
- 🔉 Codec: SNAC 24kHz (hubertsiuzdak/snac_24khz)
- 🌍 Language: Hmong (Hmoob / Hmong Daw)
- 🧠 Finetuned using Unsloth PEFT LoRA
- 🎙️ Supports emotion tags:
<giggle>,<laugh>,<sigh>,<cough>,<sniffle>,<groan>,<yawn>,<gasp> - 🎭 Optional multi‑speaker prompt prefix
- ⚡ Real‑time inference on a single GPU
🧪 Quick Inference Example
🚀 Start in Colab
import torch
from unsloth import FastLanguageModel
MODEL_PATH = "Pakorn2112/Orpheus-TTS-hmong-3b"
model, tokenizer = FastLanguageModel.from_pretrained(
model_name = MODEL_PATH,
max_seq_length = 2048, # Choose any for long context!
dtype = None, # Select None for auto detection
load_in_4bit = False, # Select True for 4bit which reduces memory usage
# token = "YOUR_HF_TOKEN", # HF Token for gated models
)
💡 Tips for Best Quality
- Use 24kHz mono WAV recordings
- Trim silence and remove heavy noise
- Keep clips 1‑8 seconds long per utterance
- Use clear, natural speaking tone
- Add optional emotion tokens for expressive voices
📄 License
apache-2.0
This model is released publicly for research & educational use. Commercial applications may require dataset rights & additional review.
🤝 Credits
- Hmong TTS Model: LocalVoice.org
- HPC Support: ThaiSC Supercomputer (LANTA) — HPC Ignite Program
- SNAC Codec Team: hubertsiuzdak (24kHz codec)
- Fine‑Tuning Framework: Unsloth
☕ Support the Project
If this model helped your project and you'd like to support future development, consider buying me a coffee.
Your support helps improve Hmong speech technology, collect better datasets, and release more open-source models for the community.
Thank you for supporting open-source Hmong AI! 🖤💚💙
🎉 Thank you for supporting Hmong language technology! 🖤💚💙