|
Download README.md from edwardgl/rumblex: direct link, hf CLI and curl.
- Browser
- Download file 856 Bytes
-
https://huggingface.co/spaces/edwardgl/rumblex/resolve/main/README.md
- Command line
-
hf download hf://spaces/edwardgl/rumblex/README.md
-
curl -L -o README.md https://huggingface.co/spaces/edwardgl/rumblex/resolve/main/README.md
856 Bytes
| title: rumblex | |
| emoji: 🎙️ | |
| colorFrom: blue | |
| colorTo: indigo | |
| sdk: gradio | |
| sdk_version: 6.32.0 | |
| python_version: "3.12" | |
| app_file: app.py | |
| pinned: false | |
| license: apache-2.0 | |
| models: | |
| - Reson8/Rumble-2-Flash | |
| short_description: Word timestamps and subtitles from Rumble-2-Flash's CTC head | |
| Demo of [rumblex](https://github.com/EdwardGlockner/rumblex): transcription with word-level timestamps and | |
| SRT/VTT/JSON output on [Reson8/Rumble-2-Flash](https://huggingface.co/Reson8/Rumble-2-Flash). Word times come | |
| from the model's own CTC head, aligned on the encoder pass the decoder already ran. | |
| This free demo transcribes the first 10 minutes of each file (2 on CPU hardware). For real use: | |
| ```bash | |
| pip install rumblex | |
| rumblex meeting.mp4 -f srt vtt --max_line_width 42 | |
| ``` | |
| Example audio: one LibriSpeech utterance (CC BY 4.0). | |