Skip to content
Audio

whisper-base-ko-coreml

корейская ASR на устройстве

1 min readupdated

Корейский файнтюн openai/whisper-base от okdohyuk, сконвертированный в Core ML для WhisperKit. Это полный файнтюн для локальной корейской транскрипции; модель используется приложением Nogmi на Apple-устройствах.

Ключевые особенности

  • Обучение на FLEURS, Zeroth-Korean и Mozilla Common Voice 17.0 (ko).
  • Конвертация через whisperkittools; папка nogmi_whisper-base-ko-v1/.
  • CER на fleurs_ko_test: 6.357% против 16.176% у базовой модели.
  • WER на fleurs_ko_test: 17.015% против 33.115%; на cv17_ko_test CER 13.126% против 28.566%.
  • Английский и японский сильно просели: CER 111.152% и 108.389% соответственно.

Пример

Сценарий из карточки: Nogmi использует nogmi_whisper-base-ko-v1/ для on-device корейской транскрипции через WhisperKit на Apple-устройствах. Отдельного кода запуска в README нет.

ПодробнееОтложенные наборы: cv17_ko_test CER 13.126%, zeroth_ko_test CER 8.135%.
Core ML vs PyTorch: Δ CER +0.0 п.п. на fleurs_ko_test и cv17_ko_test.
Длинные записи (встречи) не проверялись: внутреннего набора не было.
Веса — Apache-2.0; данные — FLEURS, Zeroth-Korean и Common Voice 17.0.
Discuss Telegram 0 14

Will it run?

Runs in BF16

Fits without quantisation.

  • BF16weights 0.1 GB · ready files0.6 GB
  • FP8 / INT8recommendedweights 0.0 GB0.6 GB
  • 4-bitweights 0.0 GB0.5 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 · BF16 — 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.

    # okdohyuk/whisper-base-ko-coreml
    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 okdohyuk/whisper-base-ko-coreml --local-dir ~/models/whisper-base-ko-coreml
    # The model is in ~/models. How to run it is in the model card (link above).
    Add more LoRAs to whisper-base-ko-coreml
  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 whisper-base (quants, finetunes, LoRAs) will show up under the bell.

Quantised versions 12All on Hugging Face
Fine-tunes 12All on Hugging Face
LoRAs and adapters 1All on Hugging Face

About the model

okdohyuk/whisper-base-ko-coreml
Author
okdohyuk
Category
Audio
Released
Posted

On NeyroLearn

Hugging Face

140

Ask AI about this model

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

Comments