Skip to content
Audio

VoxPocket-v5

компактный кхмерский TTS с клонированием голоса

1 min readupdated

VoxPocket-v5 — TTS-модель на 0,89B параметров для кхмерского языка, построенная на VoxCPM 1.5. Она читает кхмерский, кхмерский с английскими словами и простой английский, умеет клонировать голос по короткому фрагменту и принимать стиль или описание голоса. Аудио — 44,1 кГц.

Ключевые особенности

  • Клонирование голоса по чистому клипу 4–9 с и точной расшифровке сказанного.
  • Стили в скобках: шёпот, злость, радость, удивление, грусть и другие.
  • Можно задать голос описанием, без референсного аудио.
  • По бенчмарку автора читает кхмерский лучше VoxCPM2 (2,38B), хотя сама меньше.
  • Поддержка кхмерского, кхмерского с английскими словами и простого английского.

Пример

from voxcpm.core import VoxCPM
import soundfile as sf

m = VoxCPM.from_pretrained(hf_model_id="INGMUYLEANG/VoxPocket-v5", load_denoiser=False)
wav = m.generate(text="(A young woman, clear and bright voice, cheerful tone)សួស្តីបងប្អូនទាំងអស់គ្នា។")
sf.write("out.wav", wav, m.tts_model.sample_rate)
ПодробнееМедианная ошибка букв: 4,3% на 40 отложенных фразах; у VoxCPM2 — 4,8%.
На RTX 5090: 3,6× реального времени, пик VRAM 3,8/6,3 ГБ для короткого/длинного текста, 173 Вт.
Ограничения: длинный текст лучше делить на предложения; английский базовый; стили на клонированном голосе меняют тон слабее; не использовать для имитации реальных людей.
Примеры
Discuss Telegram 1 25

Will it run?

Runs in F32

Fits without quantisation.

  • F32weights 3.0 GB · ready files3.8 GB
  • BF16weights 1.5 GB2.1 GB
  • FP8 / INT8recommendedweights 0.7 GB1.3 GB
  • 4-bitweights 0.4 GB1.0 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 2060

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.

    # INGMUYLEANG/VoxPocket-v5
    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 INGMUYLEANG/VoxPocket-v5 --local-dir ~/models/VoxPocket-v5
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to VoxPocket-v5
  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. MiniCPM4-0.5Bfamily root
  2. VoxCPM1.5
  3. VoxPocket-v5

New posts about versions of MiniCPM4-0.5B (quants, finetunes, LoRAs) will show up under the bell.

Quantised versions 12All on Hugging Face
Fine-tunes 12All on Hugging Face
LoRAs and adapters 4All on Hugging Face

About the model

INGMUYLEANG/VoxPocket-v5
Author
INGMUYLEANG
Category
Audio
Released
Posted

On NeyroLearn

Hugging Face

251

Ask AI about this model

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

Comments