romara-labs/Qwen3-TTS-12Hz-1.7B-Base-ONNX

By romara-labs · ↓ 0 · ♥ 0 · Updated 2026-03-15

What's in this repo

This page lists the files in romara-labs/Qwen3-TTS-12Hz-1.7B-Base-ONNX, including code_predictor.onnx, speaker_encoder.onnx, talker_decode.onnx and more.

5 ONNX models (.onnx files)
39 files, including 22 .npy, 5 .onnx, 4 .data, 4 .json, 1 .md, 1 .py
Last updated 2026-03-15

ONNX files

File	Size	Download
code_predictor.onnx	—	Download
speaker_encoder.onnx	—	Download
talker_decode.onnx	—	Download
talker_prefill.onnx	—	Download
vocoder.onnx	—	Download

How to use

These ONNX files can be loaded with ONNX Runtime. To inspect a model graph, upload the file to the Netron viewer. Download links point to the original files on Hugging Face.

README

🎙️ Qwen3-TTS-12Hz-1.7B-Base (ONNX)

🚀 Overview

Qwen3-TTS-12Hz-1.7B-Base-ONNX is the optimization of the Qwen3-TTS framework. This model implements a discrete multi-codec Language Model (LM) architecture capable of 3-second rapid voice cloning with enhanced prosody and vocal fidelity.

The ONNX conversion enables low-latency, cross-platform deployment on both high-end CPUs and NVIDIA GPUs.

💎 Key Features

Zero-Shot Voice Cloning: High-similarity cloning (>97%) using only 3 seconds of reference audio.
Ultra-Low Latency: End-to-end streaming generation as low as 97ms.
Decoupled Architecture: Separate components for text processing, token generation, and speech synthesis.
Multilingual Excellence: Native-level pronunciation for 10 major global languages.
Vocal Richness: 2048-dimensional speaker embeddings for superior similarity.

🏗️ Model Architecture

A complex modular pipeline consisting of:

Talker (Transformer): 28 layers (Hidden Size: 2048, 8 KV Heads).
Code Predictor: 5-layer Transformer for multi-codec resolution.
Vocoder: BigVGAN-based high-fidelity speech decoder.
Speaker Encoder: ECAPA-TDNN for embedding extraction.

📦 Model Components (Modular Specs)

Component	File	Description	Output
Talker Prefill	`talker_prefill.onnx`	Initial text processing & KV Cache setup.	Logits & Hidden states.
Talker Decode	`talker_decode.onnx`	Iterative token generation logic.	New KV Cache.
Code Predictor	`code_predictor.onnx`	Multi-codec prediction (12Hz).	Multi-codebook codes.
Vocoder	`vocoder.onnx`	Final waveform synthesis.	24kHz Audio.
Speaker Enc.	`speaker_encoder.onnx`	Reference audio analysis.	2048-dim Embedding.

🛠️ Installation

pip install onnxruntime-gpu librosa soundfile numpy torch transformers

Model info

text-to-speech onnxruntime

onnxruntime onnx text-to-speech voice-cloning cpu-inference qwen3-tts base_model:Qwen/Qwen3-TTS-12Hz-1.7B-Base base_model:quantized:Qwen/Qwen3-TTS-12Hz-1.7B-Base license:apache-2.0 region:us

↓ 0 ♥ 0

Created: 2026-03-09

Updated: 2026-03-15

View on Hugging Face

Files (39)

.gitattributes

README.md

code_predictor.onnx ONNX

code_predictor.onnx.data

embeddings/codec_head_weight.npy

embeddings/config.json

embeddings/cp_codec_embedding_0.npy

embeddings/cp_codec_embedding_1.npy

embeddings/cp_codec_embedding_10.npy

embeddings/cp_codec_embedding_11.npy

embeddings/cp_codec_embedding_12.npy

embeddings/cp_codec_embedding_13.npy

embeddings/cp_codec_embedding_14.npy

embeddings/cp_codec_embedding_2.npy

embeddings/cp_codec_embedding_3.npy

embeddings/cp_codec_embedding_4.npy

embeddings/cp_codec_embedding_5.npy

embeddings/cp_codec_embedding_6.npy

embeddings/cp_codec_embedding_7.npy

embeddings/cp_codec_embedding_8.npy

embeddings/cp_codec_embedding_9.npy

embeddings/speaker_ids.json

embeddings/talker_codec_embedding.npy

embeddings/text_embedding.npy

embeddings/text_projection_fc1_bias.npy

embeddings/text_projection_fc1_weight.npy

embeddings/text_projection_fc2_bias.npy

embeddings/text_projection_fc2_weight.npy

infer_onnx.py

merges.txt

speaker_encoder.onnx ONNX

talker_decode.onnx ONNX

talker_decode.onnx.data

talker_prefill.onnx ONNX

talker_prefill.onnx.data

tokenizer_config.json

vocab.json

vocoder.onnx ONNX

vocoder.onnx.data

Related models

More text-to-speech models

Browse all text-to-speech models