SiangTTS — Thai Voice-Cloning TTS (VoxCPM2 LoRA)
SiangTTS (เสียง =
voice) is a LoRA adapter for
openbmb/VoxCPM2 that gives it clear, natural
Thai speech
with zero-shot
voice cloning, trained on a single RTX 3090 (24 GB).
It's a parameter-efficient (LoRA) fine-tune of VoxCPM2 — small enough to train on
one consumer GPU while keeping the base model's voice-design and cloning abilities.
The Thai-adaptation approach is inspired by JaiTTS (a separate, closed-source
Thai VoxCPM model by others); SiangTTS is an independent open reproduction at LoRA
scale and is not affiliated with it.
Results
Measured with Typhoon-Whisper-Large-v3 (Thai ASR; CER) and WavLM x-vectors
(speaker SIM) on small eval sets, so numbers are directional. CER is an upper
bound on error — the ASR judge itself mis-recognises some rare/archaic Thai
words the model pronounces correctly.
VoxCPM2's base is already a capable Thai speaker (it reads numerals and handles
long-form). SiangTTS clones at essentially the real same-speaker similarity
ceiling and is as intelligible as the original recordings, with much lower CER
than the base. The GT column is the real recording, included as the reference.
| Voice cloning (80 prompts) | GT (real rec.) | Base | SiangTTS |
|---|
| Intelligibility — CER ↓ | 0.97% | 3.26% | 0.84% |
| Speaker similarity — SIM ↑ | 0.913 | 0.906 | 0.909 |
(CER ≤ GT reflects clean, ASR-friendly synthesis plus the ASR judge's own floor;
SIM ≈ GT means cloning is about as close as two real recordings of one speaker.)
Separately, short-form Thai CER 5.7%→3.8% and long-form 2.7%→1.6% (small 5 / 2
-prompt sets — directional).
Trained 2 epochs over ~205 h: Common Voice Thai (diverse speakers) +
porjai_central (studio-clean) + a LibriTTS-R English slice (retains English &
code-switching). Audio encoded at 16 kHz, generated at 48 kHz (VoxCPM2 design).
Usage
1from voxcpm import VoxCPM
2from voxcpm.model.voxcpm2 import LoRAConfig
3import json
4
5cfg = json.load(open("lora_config.json"))["lora_config"]
6model = VoxCPM.from_pretrained(
7 "openbmb/VoxCPM2",
8 lora_config=LoRAConfig(**cfg),
9 lora_weights_path=".", # dir holding lora_weights.safetensors
10)
11
12# Plain TTS
13wav = model.generate(text="สวัสดีครับ ยินดีที่ได้รู้จัก", cfg_value=2.5, inference_timesteps=10)
14
15# Voice cloning from a 3-10 s reference clip
16wav = model.generate(text="ทดสอบการโคลนเสียง", reference_wav_path="ref.wav",
17 cfg_value=2.5, inference_timesteps=10)
Or via the CLI: voxcpm clone --text "..." --reference-audio ref.wav --lora-path . -o out.wav
Limitations
- Rare archaic / liturgical Pali-Sanskrit vocabulary may occasionally be
mispronounced (no phonetic-respelling dict is used — raw-text approach).
- Eval prompt sets are small; numbers are directional.
- Best for everyday/conversational Thai; not tuned for specific domains.
License
cc-by-sa-4.0 - inherited from the most restrictive training-data license
(porjai_central is CC-BY-SA-4.0; Common Voice is CC0; LibriTTS-R is CC-BY-4.0).
Commercial use is permitted under share-alike. The base model
openbmb/VoxCPM2 is Apache-2.0; SiangTTS code is Apache-2.0.
