Robust Speech Recognition via Large-Scale Weak Supervision
Open model that transcribes and translates audio files on your own hardware; no managed service, streaming or speaker labels.
Alternativen zu
Managed speech recognition service that transcribes audio files and live streams, with speaker labels and custom vocabularies. Ein geschlossenes Produkt von Amazon Web Services.
Whisper, WhisperX, Vosk und weitere. 5 Alternativen zu Amazon Transcribe, in Python, Jupyter Notebook. Ersatzgrad, Lizenz, Sterne und das neueste signierte Release im Vergleich.
| Tool | Ersatzgrad | Sterne | Lizenz | Bedingungen | Selbst hostbar | Sprache | Neuestes Release | Letzter Push |
|---|---|---|---|---|---|---|---|---|
| Whisper | Teilweise | 111k | MIT | Open Source | Ja | Python | v20250625 | 2026-08-31 |
| WhisperX | Teilweise | 25k | BSD-2-Clause | Open Source | Ja | Python | v3.8.6 | 2026-10-18 |
| Vosk | Teilweise | 15k | Apache-2.0 | Open Source | Ja | Jupyter Notebook | v0.3.50 | 2026-10-15 |
| WhisperLive | Teilweise | 4.3k | MIT | Open Source | Ja | Python | v0.10.0 | 2026-10-07 |
| Speaches | Teilweise | 3.7k | MIT | Open Source | Ja | Python | v0.9.0-rc.3 | 2026-10-17 |
5 Alternativen
Robust Speech Recognition via Large-Scale Weak Supervision
Open model that transcribes and translates audio files on your own hardware; no managed service, streaming or speaker labels.
WhisperX: Automatic Speech Recognition with Word-level Timestamps (& Diarization)
Batch transcription with speaker labels on your own hardware; no managed service or streaming.
Offline speech recognition API for Android, iOS, Raspberry Pi and servers with Python, Java, C# and Node
Offline speech recognition on your own device, without speaker labels or a managed service.
A nearly-live implementation of OpenAI's Whisper.
Near-live transcription of microphone or stream audio over WebSocket, run on your own server.
OpenAI API-compatible server for streaming transcription with faster-whisper, run on your own hardware.