Alt Denoiser vs DeepFilterNet
Alt Denoiser
github.comDeepFilterNet
github.comSide by side
| Figure | Alt Denoiser | DeepFilterNet |
|---|---|---|
| Price from | — | Free |
| Free plan | — | Yes |
| Free trial | — | — |
| Open source | — | Yes |
| Windows | Yes | Yes |
| macOS | Yes | Yes |
| Linux | Yes | Yes |
| API | — | Yes |
| Self-hosted | — | Yes |
| Plans published | 0 | 2 |
| Spec score · AI Noise Cancellation Software | 7.5 (#12) | 9.0 (#1) |
| Latency | The v1.0.2 release notes state that the processing FIFO uses a declared 1920-sample latency. | The LADSPA plugin has a minimum latency of 20 ms from STFT processing, plus latency from the LADSPA host. |
| License | The project is licensed under GPLv3. | The repository code is free and open source under either the MIT License or Apache License 2.0, at the user's option. |
| Purpose | Alt Denoiser is a real-time AI audio noise suppression plugin based on DeepFilterNet. | DeepFilterNet is a low-complexity speech enhancement framework for full-band 48 kHz audio using deep filtering. |
Plans
In their own words
Alt Denoiser
Alt Denoiser is a real-time audio noise suppression plugin based on DeepFilterNet. It is free and open source, with precompiled downloads for Windows, macOS and Linux in VST3, VST2, AU, LV2 and Standalone formats. DeepFilterNet3 models are bundled, so no separate model download is needed. The plugin accepts any host sample rate and resamples automatically. Its interface has an attenuation control and input and output RMS meters. Version 1.0.2 supports mono-to-mono, stereo-to-stereo and mono-to-stereo routing, and declares 1,920 samples of latency for startup and reset FIFO prefill. Release notes mention fixes for stutter and dropouts in EqualizerAPO and DAWs at standard host buffer sizes. Linux build instructions cover Ubuntu/Debian, Fedora and Arch; building from source requires CMake 3.15 or later, Rust and a C++ compiler. The project is licensed under GPLv3 and directs questions and bug reports to GitHub Issues. macOS binaries target macOS 14.0 and later.
DeepFilterNet
DeepFilterNet is a free, open-source framework for speech enhancement on full-band 48 kHz audio, using deep filtering. Its command-line tool suppresses noise in WAV files, with 48 kHz as its only supported sampling rate. A LADSPA plugin provides real-time noise suppression and can be used with PipeWire filter chains to create a virtual microphone or audio output, including for VoIP apps such as Zoom or Discord. The plugin's minimum latency is 20 ms from STFT processing, plus any delay added by the LADSPA host. The Rust command-line tool defaults to DeepFilterNet2; the Python interface documents DeepFilterNet and DeepFilterNet2. An optional post-filter may slightly over-attenuate very noisy sections. The framework supports Linux, macOS and Windows, though training is tested only under Linux. The Python package installs as deepfilternet, and an optional train extra adds data-loading functionality for Linux training. Code is available under either the MIT License or Apache License 2.0.