lars76 / swift-f0
Fast and accurate fundamental frequency (F0) detector using convolutional neural networks
Why this job: says: neural networks; topics: deep-learning, machine-learning
What it does
In the Pitch Detection Benchmark, SwiftF0 has the highest pitch F1 of the 19 trackers tested, statistically tied with RMVPE. It runs at 180 times real time on a single laptop CPU core. Streaming needs 176 ms of lookahead. It supports frequencies between 46.875 Hz and 2093.75 Hz (roughly F♯1 to C7). The article explains how it works and the training repository how it was trained.
Get started
pip install swift-f0- Maintenance
- Active
- Last commit 48 days ago
- Stars
- 200
- +5 this week
- Licence
- MIT
- Allows business use
- Downloads for
- None listed
- No release files
- Latest version
- v0.3.0
- Released 48 days ago
- Built with
- Python
- 100% of the code
Is it looked after?
- Contributors
- Not stated
- Last 90 days
- Issues
- 75% closed
- 1 still open
- Pull requests
- 0 merged
- 0 waiting
- Releases
- 5
- 5 in the last year
- No contributing guide
- No security policy
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| ORNL / HydraGNN | 128 | 3 days ago | BSD-3-Clause | No downloads |
| BrightSec / shainet | 198 | today | MIT | No downloads |
More: languages, topics and project record
Languages
- Python 100%
Topics
- audio-analysis
- convolutional-neural-networks
- deep-learning
- fundamental-frequency
- machine-learning
- onnx
- pitch-detection
- python
- signal-processing
- speech-processing
Project record
- JobAI libraries and models
- StatusListed
- Commits10
- Forks23
- Watching7
- Stars in 28 days+6
- CreatedJul 8, 2025
- First seen by usNov 7, 2026
- CheckedNov 8, 2026