Port of OpenAI's Whisper model in C/C++
C and C++ port of Whisper that runs on CPU and Apple silicon without Python or PyTorch.
Robust Speech Recognition via Large-Scale Weak Supervision
whisper.cpp、faster-whisper、WhisperX など。Whisperの代替4件(C++, Python, Jupyter Notebook)。置き換えの度合い、ライセンス、スター数、署名付きの最新リリースを比較できます。
| ツール | 置き換えの度合い | スター | ライセンス | 利用条件 | セルフホスト | 言語 | 最新リリース | 最終プッシュ |
|---|---|---|---|---|---|---|---|---|
| whisper.cpp | 完全な置き換え | 54k | MIT | オープンソース | 可 | C++ | v1.9.5 | 2026-10-12 |
| faster-whisper | 完全な置き換え | 26k | MIT | オープンソース | 可 | Python | v1.2.1 | 2026-10-16 |
| WhisperX | 完全な置き換え | 25k | BSD-2-Clause | オープンソース | 可 | Python | v3.8.6 | 2026-10-18 |
| Vosk | 部分的 | 15k | Apache-2.0 | オープンソース | 可 | Jupyter Notebook | v0.3.50 | 2026-10-15 |
代替4件
Port of OpenAI's Whisper model in C/C++
C and C++ port of Whisper that runs on CPU and Apple silicon without Python or PyTorch.
Faster Whisper transcription with CTranslate2
Reimplementation of Whisper on CTranslate2 that loads the same models and runs several times faster with less memory.
WhisperX: Automatic Speech Recognition with Word-level Timestamps (& Diarization)
Adds word-level timestamps, voice activity detection and speaker diarization on top of faster-whisper.
Offline speech recognition API for Android, iOS, Raspberry Pi and servers with Python, Java, C# and Node
Small offline models with streaming recognition and an API for many languages, less accurate than Whisper.