earshot
| Crates.io | earshot |
| lib.rs | earshot |
| version | 0.1.0 |
| created_at | 2024-09-29 16:39:46.007543+00 |
| updated_at | 2024-09-29 16:39:46.007543+00 |
| description | Ridiculously fast voice activity detection in pure #[no_std] Rust |
| homepage | |
| repository | https://github.com/pykeio/earshot |
| max_upload_size | |
| id | 1390980 |
| size | 71,524 |
(decahedron1)
documentation
README
Earshot
Ridiculously fast, only slightly bad voice activity detection in pure Rust. Port of the famous WebRTC VAD.
Features
#![no_std], doesn't even requirealloc- Internal buffers can get pretty big when stored on the stack, so the
allocfeature is enabled by default, which allocates them on the heap instead.
- Internal buffers can get pretty big when stored on the stack, so the
- Stupidly fast; uses only fixed-point arithmetic
- Achieves an RTF of ~3e-4 with 30 ms 48 KHz frames, ~3e-5 with 30 ms 8 KHz frames.
- Comparatively, Silero VAD v4 w/
ortachieves an RTF of ~3e-3 with 60 ms 16 KHz frames.
- Okay accuracy
- Great at distinguishing between silence and noise, but not between noise and speech.
- Earshot provides alternative models with slight accuracy gains compared to the base WebRTC model.