--- license: other language: - zh - nan - hak - en - id base_model: openai/whisper-large-v2 pipeline_tag: automatic-speech-recognition tags: - dginfra - whisper - automatic-speech-recognition - taiwanese - hakka - indonesian --- # Taiwan-Tongues-ASR-CE-pretrained Model v2.0 Taiwan-Tongues-ASR-CE-pretrained-v2.0 是以 `openai/whisper-large-v2` 為基礎訓練的 Hugging Face Transformers 版本模型。此版本新增印尼文(Indonesian, `id`)語音辨識能力,並持續支援華語、台語、客語與英語。 這個 repository 提供 `model.safetensors` 與完整 tokenizer / processor 設定,適合用於 Transformers 推論、繼續訓練或微調。若需要 CTranslate2 / faster-whisper 推論版本,請使用 [Taiwan-Tongues-ASR-CE-v2.0](https://huggingface.co/adi-gov-tw/Taiwan-Tongues-ASR-CE-v2.0)。 專案原始碼與訓練流程請見 [Taiwan-Tongues-ASR-CE](https://github.com/adi-gov-tw/Taiwan-Tongues-ASR-CE)。 ## Version | Version | Base model | Format | Main update | |---|---|---|---| | v2.0 | `openai/whisper-large-v2` | Transformers (`model.safetensors`) | Added Indonesian (`id`) support | ## Supported Languages | Language | Code | |---|---| | Mandarin Chinese | `zh` | | Taiwanese Hokkien | `nan` | | Hakka | `hak` | | English | `en` | | Indonesian | `id` | ## Files This repository contains the Transformers checkpoint and tokenizer assets: - `model.safetensors` - `config.json` - `generation_config.json` - `preprocessor_config.json` - `tokenizer.json` - `tokenizer_config.json` - `vocab.json` - `merges.txt` - `normalizer.json` - `added_tokens.json` - `special_tokens_map.json` ## Usage Install the required packages: ```bash pip install torch transformers ``` Run automatic speech recognition with Transformers: ```python from transformers import pipeline asr = pipeline( "automatic-speech-recognition", model="adi-gov-tw/Taiwan-Tongues-ASR-CE-pretrained-v2.0", device=0, ) result = asr( "audio.wav", generate_kwargs={ "language": "id", "task": "transcribe", }, ) print(result["text"]) ``` Change `language` to `zh`, `nan`, `hak`, `en`, or `id` according to the input audio. ## Notes - This is the pretrained / Transformers model package. - For faster production inference with CTranslate2 or faster-whisper, use [Taiwan-Tongues-ASR-CE-v2.0](https://huggingface.co/adi-gov-tw/Taiwan-Tongues-ASR-CE-v2.0). - The license is listed as `other`; please refer to the project and release documentation before redistribution or commercial use.
