tag:github.com,2008:https://github.com/brummer10/NeuralRack/releases 2026-01-19T05:26:18Z tag:github.com,2008:Repository/950004433/v0.3.1 2026-01-19T05:47:19Z <p>NeuralRack is a Neural Model and Impulse Response File loader for Linux/Windows available as Stand alone application, and in the Clap, LV2 and vst2 plugin format.</p> <p>It supports <a href="https://www.tone3000.com/search?tags=103" rel="nofollow">*.nam files</a> and, or <a href="https://www.tone3000.com/search?tags=23562" rel="nofollow">*.json or .aidax files</a> by using the <a href="https://github.com/mikeoliphant/NeuralAudio">NeuralAudio</a> engine.</p> <p>For Impulse Response File Convolution it use <a href="https://github.com/HiFi-LoFi/FFTConvolver">FFTConvolver</a></p> <p>Resampling is done by <a href="https://kokkinizita.linuxaudio.org/linuxaudio/zita-resampler/resampler.html" rel="nofollow">Libzita-resampler</a></p> <p>New in this release:</p> <ul> <li>Fix reload session with the clap/vst2 plugs</li> </ul> <p><a target="_blank" rel="noopener noreferrer" href="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true"><img src="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true" alt="NeuralRack" style="max-width: 100%;"></a></p> <p>Neuralrack allow to load up to two model files and run them serial.<br> The input/output could be controlled separate for each model.<br> For tone sharping a 6 band EQ could be enabled.<br> Additional it allow to load up a separate Impulse Response file for each output channel (stereo).<br> Neuralrack provide a buffered Mode which introduce a one frame latency when enabled.<br> It could move one Neural Model, or the complete processing into a background thread. That will reduce the CPU load when needed.<br> The resulting latency will be reported to the host so that it could be compensated.</p> <p>To build from source, please use NeuralRack-v0.3.1-src.tar.xz as only that contain the needed submodules.</p> github-actions[bot] tag:github.com,2008:Repository/950004433/v0.3.0 2026-01-11T03:45:12Z <p>NeuralRack is a Neural Model and Impulse Response File loader for Linux/Windows available as Stand alone application, and in the Clap, LV2 and vst2 plugin format.</p> <p>It supports <a href="https://www.tone3000.com/search?tags=103" rel="nofollow">*.nam files</a> and, or <a href="https://www.tone3000.com/search?tags=23562" rel="nofollow">*.json or .aidax files</a> by using the <a href="https://github.com/mikeoliphant/NeuralAudio">NeuralAudio</a> engine.</p> <p>For Impulse Response File Convolution it use <a href="https://github.com/HiFi-LoFi/FFTConvolver">FFTConvolver</a></p> <p>Resampling is done by <a href="https://kokkinizita.linuxaudio.org/linuxaudio/zita-resampler/resampler.html" rel="nofollow">Libzita-resampler</a></p> <p>New in this release:</p> <ul> <li>implement option to move (drag and drop) EQ around</li> </ul> <p><a target="_blank" rel="noopener noreferrer" href="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true"><img src="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true" alt="NeuralRack" style="max-width: 100%;"></a></p> <p>Neuralrack allow to load up to two model files and run them serial.<br> The input/output could be controlled separate for each model.<br> For tone sharping a 6 band EQ could be enabled.<br> Additional it allow to load up a separate Impulse Response file for each output channel (stereo).<br> Neuralrack provide a buffered Mode which introduce a one frame latency when enabled.<br> It could move one Neural Model, or the complete processing into a background thread. That will reduce the CPU load when needed.<br> The resulting latency will be reported to the host so that it could be compensated.</p> <p>To build from source, please use NeuralRack-v0.3.0-src.tar.xz as only that contain the needed submodules.</p> github-actions[bot] tag:github.com,2008:Repository/950004433/v0.2.0 2025-07-30T02:48:17Z <p>NeuralRack is a Neural Model and Impulse Response File loader for Linux/Windows available as Stand alone application, and in the Clap, LV2 and vst2 plugin format.</p> <p>It supports <a href="https://www.tone3000.com/search?tags=103" rel="nofollow">*.nam files</a> and, or <a href="https://www.tone3000.com/search?tags=23562" rel="nofollow">*.json or .aidax files</a> by using the <a href="https://github.com/mikeoliphant/NeuralAudio">NeuralAudio</a> engine.</p> <p>For Impulse Response File Convolution it use <a href="https://github.com/HiFi-LoFi/FFTConvolver">FFTConvolver</a></p> <p>Resampling is done by <a href="https://kokkinizita.linuxaudio.org/linuxaudio/zita-resampler/resampler.html" rel="nofollow">Libzita-resampler</a></p> <p>New in this release:</p> <ul> <li>implement Mix mode for IR convolver.</li> <li>implement support for ASIO control panel (Windows)</li> </ul> <p><a target="_blank" rel="noopener noreferrer" href="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true"><img src="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true" alt="NeuralRack" style="max-width: 100%;"></a></p> <p>Neuralrack allow to load up to two model files and run them serial.<br> The input/output could be controlled separate for each model.<br> For tone sharping a 6 band EQ could be enabled.<br> Additional it allow to load up a separate Impulse Response file for each output channel (stereo).<br> Neuralrack provide a buffered Mode which introduce a one frame latency when enabled.<br> It could move one Neural Model, or the complete processing into a background thread. That will reduce the CPU load when needed.<br> The resulting latency will be reported to the host so that it could be compensated.</p> <p>To build from source, please use NeuralRack-v0.2.0-src.tar.xz as only that contain the needed submodules.</p> github-actions[bot] tag:github.com,2008:Repository/950004433/v0.1.19 2025-06-10T17:24:41Z <p>NeuralRack is a Neural Model and Impulse Response File loader for Linux/Windows available as Stand alone application, and in the Clap, LV2 and vst2 plugin format.</p> <p>It supports <a href="https://www.tone3000.com/search?tags=103" rel="nofollow">*.nam files</a> and, or <a href="https://www.tone3000.com/search?tags=23562" rel="nofollow">*.json or .aidax files</a> by using the <a href="https://github.com/mikeoliphant/NeuralAudio">NeuralAudio</a> engine.</p> <p>For Impulse Response File Convolution it use <a href="https://github.com/HiFi-LoFi/FFTConvolver">FFTConvolver</a></p> <p>Resampling is done by <a href="https://kokkinizita.linuxaudio.org/linuxaudio/zita-resampler/resampler.html" rel="nofollow">Libzita-resampler</a></p> <p>New in this release:</p> <ul> <li>fix issue with higher Sample Rate based Models (resampling)</li> <li>implement Parameter handling with the Host in the Clap plugin format.</li> </ul> <p><a target="_blank" rel="noopener noreferrer" href="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true"><img src="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true" alt="NeuralRack" style="max-width: 100%;"></a></p> <p>Neuralrack allow to load up to two model files and run them serial.<br> The input/output could be controlled separate for each model.<br> For tone sharping a 6 band EQ could be enabled.<br> Additional it allow to load up a separate Impulse Response file for each output channel (stereo).<br> Neuralrack provide a buffered Mode which introduce a one frame latency when enabled.<br> It could move one Neural Model, or the complete processing into a background thread. That will reduce the CPU load when needed.<br> The resulting latency will be reported to the host so that it could be compensated.</p> <p>To build from source, please use NeuralRack-v0.1.9-src.tar.xz as only that contain the needed submodules.</p> github-actions[bot] tag:github.com,2008:Repository/950004433/v0.1.8 2025-05-28T07:18:33Z <p>NeuralRack is a Neural Model and Impulse Response File loader for Linux/Windows.</p> <p>It supports <a href="https://www.tone3000.com/search?tags=103" rel="nofollow">*.nam files</a> and, or <a href="https://www.tone3000.com/search?tags=23562" rel="nofollow">*.json or .aidax files</a> by using the <a href="https://github.com/mikeoliphant/NeuralAudio">NeuralAudio</a> engine.</p> <p>For Impulse Response File Convolution it use <a href="https://github.com/HiFi-LoFi/FFTConvolver">FFTConvolver</a></p> <p>Resampling is done by <a href="https://kokkinizita.linuxaudio.org/linuxaudio/zita-resampler/resampler.html" rel="nofollow">Libzita-resampler</a></p> <p>New in this release:</p> <ul> <li>add vst2 support</li> <li>fix issue with Clap GUI in Bitwig</li> <li>fix issue with hyper accurate models</li> <li>fix IR-file not applied to right channel</li> </ul> <p><a target="_blank" rel="noopener noreferrer" href="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true"><img src="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true" alt="NeuralRack" style="max-width: 100%;"></a></p> <p>Neuralrack allow to load up to two model files and run them serial.<br> The input/output could be controlled separate for each model.<br> For tone sharping a 6 band EQ could be enabled.<br> Additional it allow to load up a separate Impulse Response file for each output channel (stereo).<br> Neuralrack provide a buffered Mode which introduce a one frame latency when enabled.<br> It could move one Neural Model, or the complete processing into a background thread. That will reduce the CPU load when needed.<br> The resulting latency will be reported to the host so that it could be compensated.</p> <p>To build from source, please use NeuralRack-v0.1.8-src.tar.xz as only that contain the needed submodules.</p> github-actions[bot] tag:github.com,2008:Repository/950004433/v0.1.7 2025-05-20T12:49:11Z <p>NeuralRack is a Neural Model and Impulse Response File loader for Linux/Windows.</p> <p>It supports <a href="https://www.tone3000.com/search?tags=103" rel="nofollow">*.nam files</a> and, or <a href="https://www.tone3000.com/search?tags=23562" rel="nofollow">*.json or .aidax files</a> by using the <a href="https://github.com/mikeoliphant/NeuralAudio">NeuralAudio</a> engine.</p> <p>For Impulse Response File Convolution it use <a href="https://github.com/HiFi-LoFi/FFTConvolver">FFTConvolver</a></p> <p>Resampling is done by <a href="https://kokkinizita.linuxaudio.org/linuxaudio/zita-resampler/resampler.html" rel="nofollow">Libzita-resampler</a></p> <p>New in this release:</p> <ul> <li>add CLAP support</li> <li>add option to handle presets in stand-alone version</li> </ul> <p><a target="_blank" rel="noopener noreferrer" href="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true"><img src="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true" alt="NeuralRack" style="max-width: 100%;"></a></p> <p>Neuralrack allow to load up to two model files and run them serial.<br> The input/output could be controlled separate for each model.<br> For tone sharping a 6 band EQ could be enabled.<br> Additional it allow to load up a separate Impulse Response file for each output channel (stereo).<br> Neuralrack provide a buffered Mode which introduce a one frame latency when enabled.<br> It could move one Neural Model, or the complete processing into a background thread. That will reduce the CPU load when needed.<br> The resulting latency will be reported to the host so that it could be compensated.</p> <p>To build from source, please use NeuralRack-v0.1.7-src.tar.xz as only that contain the needed submodules.</p> github-actions[bot] tag:github.com,2008:Repository/950004433/v0.1.6 2025-04-26T06:12:48Z <p>NeuralRack is a Neural Model and Impulse Response File loader for Linux/Windows.</p> <p>It supports <a href="https://www.tone3000.com/search?tags=103" rel="nofollow">*.nam files</a> and, or <a href="https://www.tone3000.com/search?tags=23562" rel="nofollow">*.json or .aidax files</a> by using the <a href="https://github.com/mikeoliphant/NeuralAudio">NeuralAudio</a> engine.</p> <p>For Impulse Response File Convolution it use <a href="https://github.com/HiFi-LoFi/FFTConvolver">FFTConvolver</a></p> <p>Resampling is done by <a href="https://kokkinizita.linuxaudio.org/linuxaudio/zita-resampler/resampler.html" rel="nofollow">Libzita-resampler</a></p> <p>This release introduce a modernised UX.</p> <p><a target="_blank" rel="noopener noreferrer" href="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true"><img src="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true" alt="NeuralRack" style="max-width: 100%;"></a></p> <p>Neuralrack allow to load up to two model files and run them serial.<br> The input/output could be controlled separate for each model.<br> For tone sharping a 6 band EQ could be enabled.<br> Additional it allow to load up a separate Impulse Response file for each output channel (stereo).<br> Neuralrack provide a buffered Mode which introduce a one frame latency when enabled.<br> It could move one Neural Model, or the complete processing into a background thread. That will reduce the CPU load when needed.<br> The resulting latency will be reported to the host so that it could be compensated.</p> <p>To build from source, please use NeuralRack-v0.1.6-src.tar.xz as only that contain the needed submodules.</p> github-actions[bot] tag:github.com,2008:Repository/950004433/v0.1.5 2025-04-24T06:48:55Z <p>NeuralRack is a Neural Model and Impulse Response File loader for Linux/Windows.</p> <p>It supports <a href="https://www.tone3000.com/search?tags=103" rel="nofollow">*.nam files</a> and, or <a href="https://www.tone3000.com/search?tags=23562" rel="nofollow">*.json or .aidax files</a> by using the <a href="https://github.com/mikeoliphant/NeuralAudio">NeuralAudio</a> engine.</p> <p>For Impulse Response File Convolution it use <a href="https://github.com/HiFi-LoFi/FFTConvolver">FFTConvolver</a></p> <p>Resampling is done by <a href="https://kokkinizita.linuxaudio.org/linuxaudio/zita-resampler/resampler.html" rel="nofollow">Libzita-resampler</a></p> <p>This release implement a Noise Gate in the chain, and introduce a new rack colour scheme contributed by <a class="user-mention notranslate" data-hovercard-type="user" data-hovercard-url="/users/crshrprt/hovercard" data-octo-click="hovercard-link-click" data-octo-dimensions="link_type:self" href="https://github.com/crshrprt">@crshrprt</a></p> <p><a target="_blank" rel="noopener noreferrer" href="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true"><img src="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true" alt="NeuralRack" style="max-width: 100%;"></a></p> <p>Neuralrack allow to load up to two model files and run them serial.<br> The input/output could be controlled separate for each model.<br> For tone sharping a 6 band EQ could be enabled.<br> Additional it allow to load up a separate Impulse Response file for each output channel (stereo).<br> Neuralrack provide a buffered Mode which introduce a one frame latency when enabled.<br> It could move one Neural Model, or the complete processing into a background thread. That will reduce the CPU load when needed.<br> The resulting latency will be reported to the host so that it could be compensated.</p> <p>To build from source, please use NeuralRack-v0.1.5-src.tar.xz as only that contain the needed submodules.</p> github-actions[bot] tag:github.com,2008:Repository/950004433/v0.1.4 2025-04-02T09:44:52Z <p>NeuralRack is a Neural Model and Impulse Response File loader for Linux/Windows.</p> <p>It supports <a href="https://www.tone3000.com/search?tags=103" rel="nofollow">*.nam files</a> and, or <a href="https://www.tone3000.com/search?tags=23562" rel="nofollow">*.json or .aidax files</a> by using the <a href="https://github.com/mikeoliphant/NeuralAudio">NeuralAudio</a> engine.</p> <p>For Impulse Response File Convolution it use <a href="https://github.com/HiFi-LoFi/FFTConvolver">FFTConvolver</a></p> <p>Resampling is done by <a href="https://kokkinizita.linuxaudio.org/linuxaudio/zita-resampler/resampler.html" rel="nofollow">Libzita-resampler</a></p> <p><a target="_blank" rel="noopener noreferrer" href="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true"><img src="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true" alt="NeuralRack" style="max-width: 100%;"></a></p> <p>Neuralrack allow to load up to two model files and run them serial.<br> The input/output could be controlled separate for each model.<br> For tone sharping a 6 band EQ could be enabled.<br> Additional it allow to load up a separate Impulse Response file for each output channel (stereo).<br> Neuralrack provide a buffered Mode which introduce a one frame latency when enabled.<br> It could move one Neural Model, or the complete processing into a background thread. That will reduce the CPU load when needed.<br> The resulting latency will be reported to the host so that it could be compensated.</p> <p>To build from source, please use NeuralRack-v0.1.4-src.tar.xz as only that contain the needed submodules.</p> github-actions[bot] tag:github.com,2008:Repository/950004433/v0.1.3 2025-03-26T15:52:55Z <p>NeuralRack</p> <p>Quick Fix: Fix LV2 crash host</p> <p><a target="_blank" rel="noopener noreferrer" href="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true"><img src="https://github.com/brummer10/NeuralRack/raw/main/NeuralRack.png?raw=true" alt="NeuralRack" style="max-width: 100%;"></a></p> <p>Neuralrack allow to load up to two model files and run them serial.<br> The input/output could be controlled separate for each model.<br> For tone sharping a 6 band EQ could be enabled.<br> Additional it allow to load up a separate Impulse Response file for each output channel (stereo).<br> Neuralrack provide a buffered Mode which introduce a one frame latency when enabled.<br> It moves the complete processing into a background thread. That will reduce the CPU load when needed.<br> The resulting latency will be reported to the host so that it could be compensated.</p> <p>To build from source, please use NeuralRack-v0.1.3-src.tar.xz as only that contain the needed submodules.</p> github-actions[bot]