| title: Paragraph Inference for ASR | |
| emoji: 📝 | |
| colorFrom: blue | |
| colorTo: gray | |
| sdk: static | |
| app_file: index.html | |
| pinned: false | |
| license: mit | |
| models: | |
| - danielrosehill/MiniLM-L12-ASR-Paragraph-Breaks | |
| - danielrosehill/Gemma-3-270M-ASR-Paragraph-Breaks | |
| - danielrosehill/Gemma-3-270M-ASR-Paragraph-Breaks-ONNX | |
| datasets: | |
| - danielrosehill/ASR-Paragraph-Breaks | |
| short_description: What paragraph inference is, and where it fits in ASR | |
| # Paragraph inference for ASR | |
| One-page explainer for the ASR paragraph-break models: the task, why it gets its own small model, | |
| the intended pipelines (diagram), the models and prior art. Static HTML (`index.html`), no build step. | |
| Source: [`explainer/`](https://github.com/danielrosehill/ASR-Paragraph-Breaks/tree/main/explainer). | |