# kotoba-tech/kotoba-whisper-v2.0

Model from Hugging Face for automatic speech recognition: kotoba-tech/kotoba-whisper-v2.0 · transformers

## Facts

- Type: Audio
- Task: automatic-speech-recognition
- Framework: transformers
- License: apache-2.0 — Permissive open license: commercial use, modification, and redistribution are permitted. (https://spdx.org/licenses/Apache-2.0.html)
- Source: https://huggingface.co/kotoba-tech/kotoba-whisper-v2.0
- Hosting: hosted (10 files)
- Files: 12 listed, none hosted here
- File counts: "listed" is every file in the source repository; only hosted files download from Hugging Bay, and the download plan lists the files it selects for one download, so its count can differ.
- Hosted here: full count and size unknown
- Full repo upstream: 1.52 GB
- Recommended file: 1.51 GB · model.safetensors
- Download: https://huggingface.co/kotoba-tech/kotoba-whisper-v2.0
- Per-file SHA-256: available for 0 of 10 hosted files via https://huggingbay.xyz/api/artifacts/hf-model-kotoba-tech-kotoba-whisper-v2-0/files?limit=100
- Run-ready: no (Hugging Bay has not certified a run path for this model; hosted bytes are downloadable and hash-checkable only)
- Peer-assisted fallbacks: 0 verified, 0 pending review
- Downloads: 18921
- Stars: 95
- Card completeness: 96%

## Description

External Hugging Face catalog record for kotoba-tech/kotoba-whisper-v2.0. Task: automatic speech recognition. Library: transformers. Languages: ja. Datasets referenced by card metadata: japanese-asr/whisper_transcriptions.reazonspeech.all, japanese-asr/whisper_transcriptions.reazonspeech.all.wer_10.0, japanese-asr/whisper_transcriptions.reazonspeech.all.wer_10.0.vectorized. Includes 6 model-card evaluation signals. Popularity signal at import: 11,791 downloads and 95 likes.

## Model Card Facts

- Declared license: apache-2.0
- Pipeline: automatic-speech-recognition
- Library: transformers
- Task categories: automatic-speech-recognition
- Languages: ja
- Referenced datasets: japanese-asr/whisper_transcriptions.reazonspeech.all, japanese-asr/whisper_transcriptions.reazonspeech.all.wer_10.0, japanese-asr/whisper_transcriptions.reazonspeech.all.wer_10.0.vectorized
- Metrics: wer, cer
- Paper ids: 2212.04356, 2311.00430
- Model-index entries: kotoba-tech/kotoba-whisper-v2.0
- Evaluation signals: 6 model-card eval results
- File hints: 14 upstream file samples

## Evaluation Signals

- CommonVoice 8 (Japanese test set): WER = 58.8 (automatic-speech-recognition)
- CommonVoice 8 (Japanese test set): CER = 9.2 (automatic-speech-recognition)
- ReazonSpeech (held out test set): WER = 55.6 (automatic-speech-recognition)
- ReazonSpeech (held out test set): CER = 11.6 (automatic-speech-recognition)
- JSUT Basic 5000: WER = 63.7 (automatic-speech-recognition)
- JSUT Basic 5000: CER = 8.4 (automatic-speech-recognition)

## Referenced Data

- Dataset: japanese-asr/whisper_transcriptions.reazonspeech.all
- Dataset: japanese-asr/whisper_transcriptions.reazonspeech.all.wer_10.0
- Dataset: japanese-asr/whisper_transcriptions.reazonspeech.all.wer_10.0.vectorized
- Paper: 2212.04356
- Paper: 2311.00430

## File Hints

- .gitattributes (1.5 KB)
- README.md (24 KB)
- added_tokens.json (34 KB)
- config.json (1.3 KB)
- create_student_model.py (6.2 KB)
- generation_config.json (3.8 KB)
- merges.txt (482 KB)
- model.safetensors (1.4 GB) LFS
- normalizer.json (51 KB)
- preprocessor_config.json (340 B)
- special_tokens_map.json (2.1 KB)
- tokenizer.json (2.4 MB)

## Quality Checklist

- Summary: passed
- Description: passed
- Model card metadata: passed
- License: passed
- Provenance: passed
- File inventory: passed
- Hosted bytes: passed
- Scan result: passed
- Manifest signature: pending

## Safety Notes

- External catalog metadata imported from Hugging Face public API on 2026-08-30T08:02:36.119Z.
- Open license detected as apache-2.0.
- Model-index metadata found for kotoba-tech/kotoba-whisper-v2.0.
- Model card references dataset context: japanese-asr/whisper_transcriptions.reazonspeech.all, japanese-asr/whisper_transcriptions.reazonspeech.all.wer_10.0, japanese-asr/whisper_transcriptions.reazonspeech.all.wer_10.0.vectorized.
- Not locally hosted, mirrored, signed, or rescanned by Hugging Bay.
- Not a run kit: hosted, hash-checked bytes only; run readiness is not certified for this model yet.

## Actions

- Web: https://huggingbay.xyz/artifact/hf-model-kotoba-tech-kotoba-whisper-v2-0
- Files API: https://huggingbay.xyz/api/artifacts/hf-model-kotoba-tech-kotoba-whisper-v2-0/files?limit=100
- Metadata: https://huggingbay.xyz/api/artifacts/hf-model-kotoba-tech-kotoba-whisper-v2-0/metadata.json
- Distribution: https://huggingbay.xyz/api/artifacts/hf-model-kotoba-tech-kotoba-whisper-v2-0/distribution
- Manifest: not available until files are hosted and release checks pass.
- Request mirror: POST https://huggingbay.xyz/api/artifacts/hf-model-kotoba-tech-kotoba-whisper-v2-0/mirror-request
- Submit peer fallback for review: POST https://huggingbay.xyz/api/artifacts/hf-model-kotoba-tech-kotoba-whisper-v2-0/decentralized-fallbacks