Skip to content
Audio

Higgs Audio v3 Studio

локальный TTS с клонированием голоса на Windows

2 min read

Репозиторий с runtime-файлами для Higgs Audio v3 Studio — Windows-приложения на Rust/Tauri, которое запускает портированный C++/CUDA-движок Higgs Audio v3 локально. Это не оригинальный релиз модели, а GGUF-сборки, DLL CUDA-движка, чексуммы и манифест для загрузчика приложения. Upstream — Higgs Audio v3 TTS 4B от Boson AI.

Ключевые особенности

  • Локальный TTS, voice clone, продолжение речи и multi-speaker в десктопном приложении.
  • GGUF-варианты q4_k_m, q5_k, q6_k, q8_0 и bf16 — выбор между качеством и VRAM.
  • Speaker Gallery, импорт/экспорт голосов в ZIP и кэш референсов .hspkcache.
  • Локальный API: обычный TTS, клонирование, продолжение речи, WAV/MP3 и стриминг NDJSON.
  • Whisper.cpp помогает с транскриптом референса; есть диагностика VRAM и зависимостей DLL.
  • Телеметрия железа и диагностика VRAM в Command Centre приложения.

Пример

Сценарий из карточки: скачайте installer или portable-сборку, нажмите Download Engine DLLs, выберите GGUF-модель (например, q8_0), загрузите движок и модель, затем сгенерируйте WAV/MP3 с сохранённым голосом или без него.

ПодробнееVRAM: q4_k_m — 8 ГБ, q5_k — 9 ГБ, q6_k — 10 ГБ, q8_0 — 12 ГБ, bf16 — 16 ГБ.
Движок: Windows x64, NVIDIA RTX 30xx/40xx/50xx, драйвер с CUDA 13; nvcuda.dll берётся из драйвера NVIDIA и в репозиторий не входит.
Лицензия upstream — Boson Higgs TTS 3 Research and Non-Commercial License; коммерческое использование и редистрибуция требуют проверки условий.
Это runtime-репозиторий, а не оригинальный релиз весов модели.
Discuss Telegram 0 265

Will it run?

Working out the memory needs

The bot is still collecting this model's weight sizes; the estimate will appear here in a few minutes. Meanwhile you can rent a GPU with headroom.

Massed ComputeOpen Massed Compute

Run it on Massed Compute

The script downloads the model to the rented machine. How to run it is in the model card.

  1. 1. Rent a machine with a fitting GPU

    Open Massed Compute
  2. 2. Paste the script into the machine's terminal

    Connect over SSH (the Massed Compute panel shows the command) or open a terminal on the machine's desktop.

    HF_TOKEN / CIVITAI_TOKENThe script starts with token lines: paste yours between the quotes. The Hugging Face token is for gated models, the Civitai one for Civitai models.

    # robertbschuette/Higgs-Audio-v3-Studio
    mkdir -p ~/models
    curl -LsSf https://astral.sh/uv/install.sh | sh
    source "$HOME/.local/bin/env"
    uv tool install "huggingface_hub[cli]"
    hf download robertbschuette/Higgs-Audio-v3-Studio --local-dir ~/models/Higgs-Audio-v3-Studio
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to Higgs-Audio-v3-Studio
  3. 3. Use it

    The model is in ~/models. How to run it is in the model card (link above).

Stop the machine in the panel when you are done: it bills by the minute while it runs.

How to speed it up

Picked for this model type. Choose your hardware above to see what fits you exactly.

  • FlashAttention 44.0.0b40from November 18, 2026new release

    fastest exact attention on B200

    Kernels for Blackwell (beta). Exact attention without quantisation — fine for everything, LLMs included.

    How to enable: pip install flash-attn-4; vLLM and SGLang pick it up themselves.

    Earlier generations (2)
    • FlashAttention 3 — The Hopper (H100/H200) version with FP8.
    • FlashAttention 2 — The standard for RTX 30/40/50 and A100: exact attention without the memory cost of long contexts.
  • torch.compile2.15.0from October 28, 2026new release

    10–30% for free

    Compiles the model for your GPU. Slower first run, faster afterwards.

    How to enable: model = torch.compile(model); in ComfyUI the TorchCompileModel node.

Model family

New posts about versions of Higgs-Audio-v3-Studio (quants, finetunes, LoRAs) will show up under the bell.

No known relatives of this model yet. All derivatives on Hugging Face

About the model

robertbschuette/Higgs-Audio-v3-Studio
Author
robertbschuette
Category
Audio
Released
Posted

On NeyroLearn

Hugging Face

2650

Ask AI about this model

Answered by Gemini 4 · the chat is kept only in this tab

Comments