Skip to content
Audio

IndicConformer Bodo ONNX INT8

лёгкий ASR для бодо на CPU и в браузере

1 min read

Это INT8 ONNX-конверсия ASR-модели AI4Bharat IndicConformer для языка бодо (brx), подготовленная Assam AI Initiative (AAII). Оригинальная NeMo-модель экспортирована в ONNX и квантована в INT8 для лёгкого развёртывания, в том числе в браузере. В граф включён препроцессинг, а декодирование идёт через CTC.

Ключевые особенности

  • Язык — бодо (brx), вывод в деванагари.
  • End-to-end ONNX-граф включает STFT/Mel-препроцессинг, Conformer encoder и CTC head.
  • INT8-квантование ориентировано на CPU и клиентский инференс, включая браузерные приложения.
  • Для инференса не нужен полный стек NVIDIA NeMo — достаточно ONNX Runtime.
  • В репозитории: model_cpu_int8.onnx, конфиги, id2token.json и tokenizer.

Пример

Сценарий из карточки — браузерная голосовая печать на бодо: аудио подаётся в ONNX Runtime, модель сама выполняет препроцессинг и CTC-декодирование, возвращая текст на деванагари. Подходит для лёгких ASR-приложений, образовательных и accessibility-инструментов.

ПодробнееМодель — квантованная deployment-версия; возможны небольшие отличия в точности от оригинального NeMo-чекпоинта.
Используется CTC-декодирование.
Качество зависит от языка, условий записи, диктора и домена; перед продакшеном нужна проверка на целевых данных.
Лицензия оригинальной модели — CC-BY-4.0.
Discuss Telegram 1 31

Will it run?

Runs in INT8

Fits without quantisation.

  • INT8recommendedweights 0.2 GB · ready files0.7 GB
  • 4-bitweights 0.1 GB0.6 GB

Tip: go 8-bit.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: INT8 — 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.

    # aaiiaxom/indicconformer-bodo-onnx-int8
    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 aaiiaxom/indicconformer-bodo-onnx-int8 --local-dir ~/models/indicconformer-bodo-onnx-int8
    # 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

New posts about versions of indicconformer-bodo-onnx-int8 (quants, finetunes, LoRAs) will show up under the bell.

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

About the model

aaiiaxom/indicconformer-bodo-onnx-int8
Author
aaiiaxom
Category
Audio
Released
Posted

On NeyroLearn

Hugging Face

311

Ask AI about this model

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

Comments

Audio

OmniVoice

OmniVoice
Audio

AMBIT

AMBIT