CoolFace
Modelpublic

Gigsu/vocoloco-onnx

sourceHugging Faceapache-2.0updated 3mo agoView on Hugging Face
6likes
Model Card

VocoLoco — OmniVoice ONNX Models

ONNX exports of k2-fsa/OmniVoice for browser-based text-to-speech inference via ONNX Runtime Web.

Models

FileSizeDescription
omnivoice-main-split.onnx + _data_00-_042.3 GBMain TTS model (FP32, sharded)
omnivoice-main-int8.onnx586 MBMain TTS model (INT8 quantized, for mobile/low-memory)
omnivoice-decoder.onnx83 MBAudio token decoder (tokens to waveform)
omnivoice-encoder-fixed.onnx624 MBAudio encoder for voice cloning
tokenizer.json11 MBQwen2 BPE text tokenizer

Usage

These models are designed to run in the browser via VocoLoco, a fully client-side TTS application. No server required.

Architecture

  • —Backbone: Qwen3-0.6B (28 transformer layers)
  • —Audio codec: HiggsAudioV2 (8 codebooks, 24kHz output)
  • —Generation: Iterative masked diffusion (configurable 8-32 steps)
  • —Voice cloning: Zero-shot via reference audio encoding
  • —Voice design: Text-based control (gender, pitch, accent)

License

Apache 2.0 — same as the original OmniVoice model.

Attribution

Based on OmniVoice by Xiaomi Corp (k2-fsa).