Skip to content
Audio

Humaneness Voice Large 1.7

выразительная речь на пяти языках

2 min read

LAION выпустила TTS-модель Humaneness Voice Large 1.7 с 4,13 млрд уникальных базовых параметров. Это полные merged-веса: для обычного инференса адаптер подключать не нужно. Модель генерирует английский, немецкий, французский, испанский и нидерландский, а отдельный кодек возвращает аудио 48 кГц mono.

Архитектура — иерархическая MOSS Local Transformer TTS: Qwen3-family semantic transformer и local Talker. Каждый 80-мс кадр даёт 12 индексов, по одному на codebook.

Ключевые особенности

  • Управляемая подача: 40 эмоций, 15 стилей речи и 18 классов vocal bursts в интерактивном мануале; инструкции задаются отдельно от произносимого текста.
  • Два входа: text — что именно произнести, prompt — как это сыграть, с паузами, таймингом и вокальными реакциями.
  • Референсный голос: чистый фрагмент одного говорящего от 5,04 с; helper обрезает его до 14,96 с перед кодированием.
  • CFG: 1 означает guidance off; 2.5 — настройка acting showcase, а не универсальный оптимум для всех условий.
  • В мануале 455 комбинаций target/language/condition; есть machine-readable recipes и agent guide.
  • Новые материалы — CC BY 4.0; MOSS Audio Tokenizer v2 сохраняет свои upstream-условия.

Пример

Быстрый запуск из карточки:

hf download laion/Humaneness-Voice-Large-1.7 --local-dir humaneness-large17
cd humaneness-large17
pip install -r requirements.txt
python infer.py --language en --seconds 9.2 --seed 1701 --text 'I thought you had forgotten. You remembered every little detail. I am so glad you came back.' --prompt-file examples/prompt-en.txt --output example.wav
ПодробнееБенчмарки выбора чекпойнта (108 задач EN/DE, один seed, без reference, CFG 1): Content Enjoyment 6.3334, Genuineness 4.2565, Vocal Burst Blend 7.7503, composite 0.63709; mean per-clip ASR WER 0.05787%.
Это development-выбор среди семи чекпойнтов, не независимое held-out подтверждение; оценки — предсказания модели, не human listening ratings.
Корневые веса FP32; штатный загрузчик использует BF16 weights and compute. VRAM в карточке не указана.
Референс направляет identity, но не гарантирует точное сохранение голоса. Не применяйте включённый adapter к merged root повторно.
Discuss Telegram 0 94

Will it run?

Runs in F32

Fits without quantisation.

  • F32weights 15 GB · ready files17 GB
  • BF16weights 7.7 GB9.0 GB
  • FP8 / INT8recommendedweights 3.8 GB4.7 GB
  • 4-bitweights 2.2 GB2.9 GB

Tip: go 8-bit.FP8 / INT8 is within about a percent of full precision in quality, takes half the memory and runs faster on RTX 40/50 and H100. Full precision only pays off for fine-tuning.

What to get for this model

Massed ComputeOpen Massed Compute

Your own GPU: FP8 / INT8 — from RTX 2060 · F32 — from RTX 4090

Massed Compute hourly prices, November 2026. The discounted price applies with the coupon at checkout.

Estimate: weights plus working memory. Long videos, high resolutions and long contexts need more.

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.

    # laion/Humaneness-Voice-Large-1.7
    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 laion/Humaneness-Voice-Large-1.7 --local-dir ~/models/Humaneness-Voice-Large-1.7
    # The model is in ~/models. How to run it is in the model card (link above).
  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

  1. moss-tts-local-transformer-4.55b-voice-acting-v2family root
  2. moss-tts-local-transformer-4.55b-voice-acting-v2-sft3
  3. Humaneness-Voice-Large-1.2-Alpha
  4. Humaneness-Voice-Large-1.3-Alpha
  5. Humaneness-Voice-Large-1.7

New posts about versions of moss-tts-local-transformer-4.55b-voice-acting-v2 (quants, finetunes, LoRAs) will show up under the bell.

Fine-tunes 2All on Hugging Face
LoRAs and adapters 9All on Hugging Face

About the model

laion/Humaneness-Voice-Large-1.7
Author
laion
Category
Audio
Released
Posted

On NeyroLearn

Hugging Face

940

Ask AI about this model

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

Comments