--- title: audio8-TTS-0.1B-ONNX-INT8 canonical_url: "https://www.modelscope.cn/models/Audio8/audio8-TTS-0.1B-ONNX-INT8" md_url: "https://www.modelscope.cn/models/Audio8/audio8-TTS-0.1B-ONNX-INT8.md" repository: Audio8/audio8-TTS-0.1B-ONNX-INT8 last_updated: 2026-08-28 license: apache-2.0 pipeline_tag: text-to-speech tasks: - text-to-speech model_type: - arktts architectures: - ArkttsModel base_model: - Audio8/Audio8-TTS-Preview-0.1b base_model_relation: finetune library_name: - onnx - pytorch frameworks: - pytorch language: - yue - zh - nl - en - fr - de - it - ja - ko - pl - es downloads: 63 stars: 3 tags: - onnx - int8 - audio - text-to-speech - tts - voice-cloning - zero-shot - multilingual --- # audio8-TTS-0.1B-ONNX-INT8 > audio8-TTS-0.1B-ONNX-INT8 - Audio8 在 ModelScope 开源的模型。  Audio8 TTS Preview 0.1B ONNX INT8 Audio8/audio8-TTS-0.1B-ONNX-INT8 是 ModelScope 魔搭社区上的text-to-speech模型，采用 apache-2.0 许可，基于 Audio8/Audio8-TTS-Preview-0.1b 构建。 - **Repository**: Audio8/audio8-TTS-0.1B-ONNX-INT8 - **License**: apache-2.0 - **Tasks**: text-to-speech - **Base model**: Audio8/Audio8-TTS-Preview-0.1b - **Tags**: onnx, int8, audio, text-to-speech, tts, voice-cloning, zero-shot, multilingual - **Downloads**: 63 - **Stars**: 3 - **Last updated**: 2026-08-28 Source: https://www.modelscope.cn/models/Audio8/audio8-TTS-0.1B-ONNX-INT8 ---
Audio8

  Audio8 TTS Preview 0.1B ONNX INT8

**A compact multilingual TTS with zero-shot voice cloning, packaged for low-resource CPU inference.** [![GitHub](https://img.shields.io/badge/GitHub-Audio8__TTS-black?style=for-the-badge&logo=github)](https://github.com/Audio8-AI/Audio8_TTS) [![Base Model](https://img.shields.io/badge/%F0%9F%A4%97%20Base%20Model-0.1B-FFD21E?style=for-the-badge)](https://huggingface.co/Audio8/Audio8-TTS-Preview-0.1b) [![Demo](https://img.shields.io/badge/Live-Demo-brightgreen?style=for-the-badge&logo=githubpages)](https://audio8-ai.github.io/Audio8_TTS/0.1B/) [![ONNX Runtime](https://img.shields.io/badge/ONNX-Runtime-005CED?style=for-the-badge&logo=onnx)](https://onnxruntime.ai/) [![License](https://img.shields.io/badge/License-Apache%202.0-blue?style=for-the-badge)](https://github.com/Audio8-AI/Audio8_TTS/blob/master/LICENSE)
Audio8 TTS Preview is a compact multilingual text-to-speech model with zero-shot voice cloning. This repository provides its CPU-oriented ONNX deployment: INT8 per-token DualAR models, an FP16 neural audio codec, the tokenizer, the packaged reference voice, and the optional FP16 encoder used to register reference voices. > **Model files only.** Inference, streaming service, and voice-registration > code live in the dedicated > [0.1B INT8 ONNX Runtime](https://github.com/Audio8-AI/Audio8_TTS/tree/master/onnx_runtime_0_1b_int8). > The older `onnx_runtime/` directory targets the incompatible 0.6B INT4 > graphs and must not be used with this model. > **Acknowledgments** > > The per-token ONNX export design (stacked recurrent interface, valid-prefix > attention) and the INT8 quantization strategy in this repository were > **adapted from** the community conversion > [Masterx/Audio8-TTS-Preview-0.1B-ONNX-INT8](https://huggingface.co/Masterx/Audio8-TTS-Preview-0.1B-ONNX-INT8). > The graphs and tooling are independently produced by the Audio8 team from > the official checkpoint and do not reuse that repository's artifacts. > Thanks to Masterx and the community for exploring ONNX Runtime conversion of > the 0.1B model. > ## Why this ONNX release | | Deployment characteristic | |---|---| | **CPU native** | ONNX Runtime `CPUExecutionProvider`; no CUDA requirement | | **Small runtime** | No PyTorch, Transformers, or Hugging Face Hub dependency after download | | **Low memory** | About 0.6 GB after loading the online sessions in our test configuration | | **Voice cloning** | Bundled FP16 codec encoder for reusable local voice profiles | | **Local service** | CLI, web UI, HTTP API, streaming PCM, and OpenAI-compatible endpoint | ### Precision and footprint | Component | Precision | |---|---| | Slow/Fast AR | INT8 (dynamic activation quantization, `MatMulInteger`) | | Codec encoder and decoder | FP16 | | Waveform output | FP32, 44.1 kHz mono | The slow AR runs as a per-token recurrent graph (attention + Mamba states passed explicitly) so the CPU workload stays tiny: about 19 ms per slow-AR token and 8 ms per fast-AR frame on our test host (8 threads). A one-time prompt prefill costs about 19 ms per prompt token. Normal synthesis loads only the Slow AR, Fast AR, and codec decoder sessions; the online model files occupy about 0.6 GB. The complete repository, including the voice-registration encoder, is about 1.0 GB. Actual memory use varies by platform and allocator behavior. ## Supported Languages

Cantonese  ·  Chinese  ·  Dutch  ·  English
French  ·  German  ·  Italian  ·  Japanese
Korean  ·  Polish  ·  Spanish

> **Preview status:** Language coverage is intentionally limited in this > release. For the best results, use one of the 11 recommended languages > above. Broader multilingual coverage and Chinese dialect support are > planned for future releases. ## Model Details Audio8 TTS uses a DualAR architecture inspired by [Fish Audio S2 Pro](https://github.com/fishaudio/fish-speech). The 0.1B model uses a Falcon-H1 hybrid slow AR (attention + Mamba) that predicts one semantic token per audio frame; the fast AR predicts the frame's codec codebooks, conditioned on the slow hidden state and preceding codebooks. | Component | Configuration | |---|---| | Main model | about 100M parameters, excluding the codec | | Slow AR | 24 layers, width 512, Falcon-H1 hybrid (8 attention heads, 2 KV heads, Mamba d_state 64) | | Fast AR | 4 layers, width 512 | | Acoustic tokens | 10 codebooks, 4,096 entries per codebook | | Codec | 44.1 kHz, 2,048 samples per model frame (~21.5 frames/s) | | Context | Up to 2,048 packed text/audio positions | | Execution provider | ONNX Runtime CPU | ## Quick Start Python 3.10 or newer is required. The runtime uses ONNX Runtime's `CPUExecutionProvider`; platform-specific ONNX Runtime wheels may be needed on devices such as Jetson. The current release is tested on macOS arm64 and Linux x86_64. ### 1. Download the code and model ```bash git clone https://github.com/Audio8-AI/Audio8_TTS.git cd Audio8_TTS/onnx_runtime_0_1b_int8 python3 -m pip install -U "huggingface_hub[cli]" hf download Audio8/audio8-TTS-0.1B-ONNX-INT8 --local-dir model bash setup.sh "$PWD/.venv/bin/python" scripts/register_default_voice.py # packaged reference -> "default" voice ``` The model files are stored at this Hugging Face repository's root. Downloading with `--local-dir model` creates the exact layout expected by the runtime: ```text model/ ├── slow_ar_int8.onnx # per-token slow AR (prefill + decode) ├── fast_ar_int8.onnx # per-token fast AR ├── codec_decoder_fp16.onnx(.data) ├── runtime_manifest.json ├── tokenizer/tokenizer.json ├── reference_codes.npy # packaged fixed reference └── registration/ ├── codec_encoder_fp16.onnx(.data) └── registration_manifest.json ``` ### 2. Generate speech ```bash bash run_infer.sh \ --voice default \ --text "这是一个零样本语音合成的中文测试，声音克隆效果怎么样？" \ --max-new-tokens 256 \ --output outputs/example.wav ``` The command writes `outputs/example.wav` and `[10, T]` codec codes to `outputs/example.npy`. ### 3. Register a reference voice Start the local service and open . The page provides text generation, voice selection, WAV playback/download, reference-voice registration, memory status, and runtime reload. The Swagger API documentation is available at . Upload a 0.5-30 second reference recording, its exact transcript, and a voice name to register a new voice. ```bash bash start_server.sh ``` The same operation is available through HTTP: ```bash curl http://127.0.0.1:8024/api/voices/register \ -F 'audio=@/absolute/path/reference.wav' \ -F 'text=The exact transcript of the reference recording.' \ -F 'name=speaker_a' \ -F 'overwrite=false' ``` ### HTTP API ```bash curl http://127.0.0.1:8024/api/tts \ -H 'Content-Type: application/json' \ -d '{"text":"Welcome to Audio8 TTS.","voice_name":"speaker_a","max_new_tokens":256}' \ -o outputs/api.wav ``` ### OpenAI-compatible API ```bash curl http://127.0.0.1:8024/v1/audio/speech \ -H 'Content-Type: application/json' \ -d '{"model":"arktts","input":"Welcome to Audio8 TTS.","voice":"speaker_a","response_format":"wav"}' \ -o outputs/openai.wav ``` See the complete [0.1B ONNX Runtime guide](https://github.com/Audio8-AI/Audio8_TTS/tree/master/onnx_runtime_0_1b_int8) for streaming output, configuration, memory management, and service controls. ## Evaluation The source Audio8 TTS Preview 0.1B checkpoint is a compact multilingual model; see the [base model card](https://huggingface.co/Audio8/Audio8-TTS-Preview-0.1b) for details. INT8 quantization can change sampled token sequences, so quality should be evaluated for each target language, voice, and deployment setting rather than assuming bit-for-bit equivalence with the source checkpoint. This repository contains the 0.1B INT8 graphs; use the dedicated runtime linked above when running them. ## Limitations and Responsible Use - This is a Preview checkpoint with limited multilingual and dialect coverage. - Very long, noisy, or incorrectly transcribed references can reduce stability and speaker similarity. - Generated speech can be misused for impersonation or misinformation. Obtain consent before cloning a voice and clearly disclose synthetic audio where appropriate. - Evaluate the model for accuracy, safety, and legal compliance before deployment. ## License and Acknowledgements The code and model weights are released under the [Apache License 2.0](https://github.com/Audio8-AI/Audio8_TTS/blob/master/LICENSE). See the upstream [NOTICE](https://github.com/Audio8-AI/Audio8_TTS/blob/master/NOTICE) for attribution details. We thank the Fish Audio team for publishing the DualAR architecture used in Fish Audio S2 Pro.