Views
No views yet
1!pip install omnivoice
2
3from omnivoice import OmniVoice
4import soundfile as sf
5
6model = OmniVoice.from_pretrained("hotdogs/omnivoice-thai")
7
8# สร้างเสียงจากข้อความ
9audio = model.generate(
10 text="สวัสดีครับ วันนี้อากาศดีมากเลย",
11 instruct="male, low pitch",
12)
13sf.write("output.wav", audio[0], 24000)k2-fsa/OmniVoice (Qwen3-0.6B, MaskGIT diffusion)torchaudio, tokenize ด้วย eustlb/higgs-audio-v2-tokenizerbatch_tokens: 2,048 (ต่อ GPU forward pass)gradient_accumulation_steps: 8 (effective batch ≈ 16,384 tokens)learning_rate: 1e-5, cosine schedule, warmup 2%max_steps: 30,000, early stop เมื่อ per-step loss < 3.0mixed_precision: fp16, attn_implementation: sdpa| เครื่องมือ | วัตถุประสงค์ |
|---|---|
| OmniVoice | TTS framework (MaskGIT) |
| PyTorch 2.8 + CUDA 13.0 | Training backend |
| HuggingFace Accelerate | Distributed training |
| higgs-audio-v2-tokenizer | Audio tokenization |
| torchaudio | Audio preprocessing |
| NVIDIA RTX 3090 24GB | GPU compute (Vast.ai) |
| Hermes Agent | Autonomous AI agent for orchestration |
1!pip install omnivoice
2
3from omnivoice import OmniVoice
4model = OmniVoice.from_pretrained("hotdogs/omnivoice-thai")
5
6# Voice cloning
7audio = model.generate(
8 text="Hello, this is a test.",
9 ref_audio="reference.wav",
10)
11
12# Voice design
13audio = model.generate(
14 text="The weather is nice today.",
15 instruct="female, high pitch, british accent",
16)Step 1747/30000 | loss=2.8775 (per-step) | lr=9.96e-06
Step 1700 | train/loss: 4.4281 (smoothed) | epoch: 1