# yeyupiaoling/Whisper-Finetune

Fine-tune the Whisper speech recognition model to support training without timestamp data, training with timestamp data, and training without speech data. Accelerate inference and support Web deployment, Windows desktop deployment, and Android deployment

Repository: https://github.com/yeyupiaoling/Whisper-Finetune
Canonical: https://ross.abutalabs.com/products/whisper-finetune
Language: C
License: Apache-2.0
License Family: permissive
Topics: asr, ctranslate2, huggingface, whisper, lora, speech-recognition, transformers, chinese, pytorch, android, web
Last push: 2026-05-08T01:23:47+00:00

## Health v2 (maintenance only)
Score: 66/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 81, release rhythm 35, longevity 87
- inputs: {"age_days": 1229, "days_push": 118, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1223, forks 223 (observed 2026-08-28T04:04:02.621152+00:00)

## What it is
A toolkit for fine-tuning OpenAI's Whisper speech recognition models using LoRA, supporting training with or without timestamps and even without speech data. It also provides accelerated inference via CTranslate2 and GGML, with deployment options for web servers, Windows desktop apps, and Android.

## Use cases
- fine-tune whisper on my own audio dataset
- improve chinese speech recognition accuracy of whisper
- train whisper with lora adapters
- convert whisper model to ctranslate2 for faster inference
- deploy whisper asr model on android
- run whisper speech-to-text in a windows desktop app
- evaluate word error rate of whisper models
- serve a fine-tuned whisper model via a web api

## When to avoid
- you just need to run Whisper as-is without customization
- you need speech recognition for languages or pipelines unrelated to Whisper
- you want a production-grade managed ASR service rather than a self-managed model

## Facets
- artifact type: library
- maturity: active
- function: speech-recognition, machine-learning, llm-training, audio-processing
- domain: speech-processing, machine-learning
- platform: python, windows
- tags: whisper, asr, fine-tuning, lora, ctranslate2, ggml, chinese, model-deployment, natural-language-processing, linux, macos, android, gpu

## Member repositories
- yeyupiaoling/Whisper-Finetune (main) score 66

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:02.621152+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:15:35.694316+00:00, confidence not recorded.
  - readme: https://github.com/yeyupiaoling/Whisper-Finetune (fetched 2026-08-28T04:04:02.621152+00:00, sha 13407bdb13ca)
- Data as of 2026-08-30T08:39:29.467469+00:00.