diff --git a/docs/_config.yml b/docs/_config.yml index 43a5e52d..175a76be 100644 --- a/docs/_config.yml +++ b/docs/_config.yml @@ -50,3 +50,7 @@ plugins: # - vendor/cache/ # - vendor/gems/ # - vendor/ruby/ + +header_pages: + - resources.md + - about.md \ No newline at end of file diff --git a/audio/eval/pred-GCN-3-stft-48k.wav b/docs/assets/audio/eval/pred-GCN-3-stft-48k.wav similarity index 100% rename from audio/eval/pred-GCN-3-stft-48k.wav rename to docs/assets/audio/eval/pred-GCN-3-stft-48k.wav diff --git a/audio/eval/pred-GCN-99-stft-16k.wav b/docs/assets/audio/eval/pred-GCN-99-stft-16k.wav similarity index 100% rename from audio/eval/pred-GCN-99-stft-16k.wav rename to docs/assets/audio/eval/pred-GCN-99-stft-16k.wav diff --git a/audio/eval/pred-GCN-99-stft-48k.wav b/docs/assets/audio/eval/pred-GCN-99-stft-48k.wav similarity index 100% rename from audio/eval/pred-GCN-99-stft-48k.wav rename to docs/assets/audio/eval/pred-GCN-99-stft-48k.wav diff --git a/audio/eval/pred-GCN-stft-16k.wav b/docs/assets/audio/eval/pred-GCN-stft-16k.wav similarity index 100% rename from audio/eval/pred-GCN-stft-16k.wav rename to docs/assets/audio/eval/pred-GCN-stft-16k.wav diff --git a/audio/eval/pred-GCN-stft-48k.wav b/docs/assets/audio/eval/pred-GCN-stft-48k.wav similarity index 100% rename from audio/eval/pred-GCN-stft-48k.wav rename to docs/assets/audio/eval/pred-GCN-stft-48k.wav diff --git a/audio/eval/pred-GRU-99-stft-16k.wav b/docs/assets/audio/eval/pred-GRU-99-stft-16k.wav similarity index 100% rename from audio/eval/pred-GRU-99-stft-16k.wav rename to docs/assets/audio/eval/pred-GRU-99-stft-16k.wav diff --git a/audio/eval/pred-GRU-99-stft-48k.wav b/docs/assets/audio/eval/pred-GRU-99-stft-48k.wav similarity index 100% rename from audio/eval/pred-GRU-99-stft-48k.wav rename to docs/assets/audio/eval/pred-GRU-99-stft-48k.wav diff --git a/audio/eval/pred-GRU-stft-16k.wav b/docs/assets/audio/eval/pred-GRU-stft-16k.wav similarity index 100% rename from audio/eval/pred-GRU-stft-16k.wav rename to docs/assets/audio/eval/pred-GRU-stft-16k.wav diff --git a/audio/eval/pred-GRU-stft-48k.wav b/docs/assets/audio/eval/pred-GRU-stft-48k.wav similarity index 100% rename from audio/eval/pred-GRU-stft-48k.wav rename to docs/assets/audio/eval/pred-GRU-stft-48k.wav diff --git a/audio/eval/pred-LSTM-99-stft-16k.wav b/docs/assets/audio/eval/pred-LSTM-99-stft-16k.wav similarity index 100% rename from audio/eval/pred-LSTM-99-stft-16k.wav rename to docs/assets/audio/eval/pred-LSTM-99-stft-16k.wav diff --git a/audio/eval/pred-LSTM-99-stft-48k.wav b/docs/assets/audio/eval/pred-LSTM-99-stft-48k.wav similarity index 100% rename from audio/eval/pred-LSTM-99-stft-48k.wav rename to docs/assets/audio/eval/pred-LSTM-99-stft-48k.wav diff --git a/audio/eval/pred-LSTM-stft-16k.wav b/docs/assets/audio/eval/pred-LSTM-stft-16k.wav similarity index 100% rename from audio/eval/pred-LSTM-stft-16k.wav rename to docs/assets/audio/eval/pred-LSTM-stft-16k.wav diff --git a/audio/eval/pred-LSTM-stft-48k.wav b/docs/assets/audio/eval/pred-LSTM-stft-48k.wav similarity index 100% rename from audio/eval/pred-LSTM-stft-48k.wav rename to docs/assets/audio/eval/pred-LSTM-stft-48k.wav diff --git a/audio/eval/pred-TCN-99-stft-16k.wav b/docs/assets/audio/eval/pred-TCN-99-stft-16k.wav similarity index 100% rename from audio/eval/pred-TCN-99-stft-16k.wav rename to docs/assets/audio/eval/pred-TCN-99-stft-16k.wav diff --git a/audio/eval/pred-TCN-99-stft-48k.wav b/docs/assets/audio/eval/pred-TCN-99-stft-48k.wav similarity index 100% rename from audio/eval/pred-TCN-99-stft-48k.wav rename to docs/assets/audio/eval/pred-TCN-99-stft-48k.wav diff --git a/audio/eval/pred-TCN-stft-16k.wav b/docs/assets/audio/eval/pred-TCN-stft-16k.wav similarity index 100% rename from audio/eval/pred-TCN-stft-16k.wav rename to docs/assets/audio/eval/pred-TCN-stft-16k.wav diff --git a/audio/eval/pred-TCN-stft-48k.wav b/docs/assets/audio/eval/pred-TCN-stft-48k.wav similarity index 100% rename from audio/eval/pred-TCN-stft-48k.wav rename to docs/assets/audio/eval/pred-TCN-stft-48k.wav diff --git a/audio/eval/pred-WaveNet-99-stft-16k.wav b/docs/assets/audio/eval/pred-WaveNet-99-stft-16k.wav similarity index 100% rename from audio/eval/pred-WaveNet-99-stft-16k.wav rename to docs/assets/audio/eval/pred-WaveNet-99-stft-16k.wav diff --git a/audio/eval/pred-WaveNet-99-stft-48k.wav b/docs/assets/audio/eval/pred-WaveNet-99-stft-48k.wav similarity index 100% rename from audio/eval/pred-WaveNet-99-stft-48k.wav rename to docs/assets/audio/eval/pred-WaveNet-99-stft-48k.wav diff --git a/audio/eval/pred-WaveNet-stft-16k.wav b/docs/assets/audio/eval/pred-WaveNet-stft-16k.wav similarity index 100% rename from audio/eval/pred-WaveNet-stft-16k.wav rename to docs/assets/audio/eval/pred-WaveNet-stft-16k.wav diff --git a/audio/eval/pred-WaveNet-stft-48k.wav b/docs/assets/audio/eval/pred-WaveNet-stft-48k.wav similarity index 100% rename from audio/eval/pred-WaveNet-stft-48k.wav rename to docs/assets/audio/eval/pred-WaveNet-stft-48k.wav diff --git a/audio/eval/target-GCN-3-stft-48k.wav b/docs/assets/audio/eval/target-GCN-3-stft-48k.wav similarity index 100% rename from audio/eval/target-GCN-3-stft-48k.wav rename to docs/assets/audio/eval/target-GCN-3-stft-48k.wav diff --git a/audio/eval/target-GCN-99-stft-16k.wav b/docs/assets/audio/eval/target-GCN-99-stft-16k.wav similarity index 100% rename from audio/eval/target-GCN-99-stft-16k.wav rename to docs/assets/audio/eval/target-GCN-99-stft-16k.wav diff --git a/audio/eval/target-GCN-99-stft-48k.wav b/docs/assets/audio/eval/target-GCN-99-stft-48k.wav similarity index 100% rename from audio/eval/target-GCN-99-stft-48k.wav rename to docs/assets/audio/eval/target-GCN-99-stft-48k.wav diff --git a/audio/eval/target-GCN-stft-16k.wav b/docs/assets/audio/eval/target-GCN-stft-16k.wav similarity index 100% rename from audio/eval/target-GCN-stft-16k.wav rename to docs/assets/audio/eval/target-GCN-stft-16k.wav diff --git a/audio/eval/target-GCN-stft-48k.wav b/docs/assets/audio/eval/target-GCN-stft-48k.wav similarity index 100% rename from audio/eval/target-GCN-stft-48k.wav rename to docs/assets/audio/eval/target-GCN-stft-48k.wav diff --git a/audio/eval/target-GRU-99-stft-16k.wav b/docs/assets/audio/eval/target-GRU-99-stft-16k.wav similarity index 100% rename from audio/eval/target-GRU-99-stft-16k.wav rename to docs/assets/audio/eval/target-GRU-99-stft-16k.wav diff --git a/audio/eval/target-GRU-99-stft-48k.wav b/docs/assets/audio/eval/target-GRU-99-stft-48k.wav similarity index 100% rename from audio/eval/target-GRU-99-stft-48k.wav rename to docs/assets/audio/eval/target-GRU-99-stft-48k.wav diff --git a/audio/eval/target-GRU-stft-16k.wav b/docs/assets/audio/eval/target-GRU-stft-16k.wav similarity index 100% rename from audio/eval/target-GRU-stft-16k.wav rename to docs/assets/audio/eval/target-GRU-stft-16k.wav diff --git a/audio/eval/target-GRU-stft-48k.wav b/docs/assets/audio/eval/target-GRU-stft-48k.wav similarity index 100% rename from audio/eval/target-GRU-stft-48k.wav rename to docs/assets/audio/eval/target-GRU-stft-48k.wav diff --git a/audio/eval/target-LSTM-99-stft-16k.wav b/docs/assets/audio/eval/target-LSTM-99-stft-16k.wav similarity index 100% rename from audio/eval/target-LSTM-99-stft-16k.wav rename to docs/assets/audio/eval/target-LSTM-99-stft-16k.wav diff --git a/audio/eval/target-LSTM-99-stft-48k.wav b/docs/assets/audio/eval/target-LSTM-99-stft-48k.wav similarity index 100% rename from audio/eval/target-LSTM-99-stft-48k.wav rename to docs/assets/audio/eval/target-LSTM-99-stft-48k.wav diff --git a/audio/eval/target-LSTM-stft-16k.wav b/docs/assets/audio/eval/target-LSTM-stft-16k.wav similarity index 100% rename from audio/eval/target-LSTM-stft-16k.wav rename to docs/assets/audio/eval/target-LSTM-stft-16k.wav diff --git a/audio/eval/target-LSTM-stft-48k.wav b/docs/assets/audio/eval/target-LSTM-stft-48k.wav similarity index 100% rename from audio/eval/target-LSTM-stft-48k.wav rename to docs/assets/audio/eval/target-LSTM-stft-48k.wav diff --git a/audio/eval/target-TCN-99-stft-16k.wav b/docs/assets/audio/eval/target-TCN-99-stft-16k.wav similarity index 100% rename from audio/eval/target-TCN-99-stft-16k.wav rename to docs/assets/audio/eval/target-TCN-99-stft-16k.wav diff --git a/audio/eval/target-TCN-99-stft-48k.wav b/docs/assets/audio/eval/target-TCN-99-stft-48k.wav similarity index 100% rename from audio/eval/target-TCN-99-stft-48k.wav rename to docs/assets/audio/eval/target-TCN-99-stft-48k.wav diff --git a/audio/eval/target-TCN-stft-16k.wav b/docs/assets/audio/eval/target-TCN-stft-16k.wav similarity index 100% rename from audio/eval/target-TCN-stft-16k.wav rename to docs/assets/audio/eval/target-TCN-stft-16k.wav diff --git a/audio/eval/target-TCN-stft-48k.wav b/docs/assets/audio/eval/target-TCN-stft-48k.wav similarity index 100% rename from audio/eval/target-TCN-stft-48k.wav rename to docs/assets/audio/eval/target-TCN-stft-48k.wav diff --git a/audio/eval/target-WaveNet-99-stft-16k.wav b/docs/assets/audio/eval/target-WaveNet-99-stft-16k.wav similarity index 100% rename from audio/eval/target-WaveNet-99-stft-16k.wav rename to docs/assets/audio/eval/target-WaveNet-99-stft-16k.wav diff --git a/audio/eval/target-WaveNet-99-stft-48k.wav b/docs/assets/audio/eval/target-WaveNet-99-stft-48k.wav similarity index 100% rename from audio/eval/target-WaveNet-99-stft-48k.wav rename to docs/assets/audio/eval/target-WaveNet-99-stft-48k.wav diff --git a/audio/eval/target-WaveNet-stft-16k.wav b/docs/assets/audio/eval/target-WaveNet-stft-16k.wav similarity index 100% rename from audio/eval/target-WaveNet-stft-16k.wav rename to docs/assets/audio/eval/target-WaveNet-stft-16k.wav diff --git a/audio/eval/target-WaveNet-stft-48k.wav b/docs/assets/audio/eval/target-WaveNet-stft-48k.wav similarity index 100% rename from audio/eval/target-WaveNet-stft-48k.wav rename to docs/assets/audio/eval/target-WaveNet-stft-48k.wav diff --git a/docs/index.md b/docs/index.md index 4faa014c..490961a7 100644 --- a/docs/index.md +++ b/docs/index.md @@ -3,22 +3,27 @@ title: Evaluating Neural Networks Architectures for Spring Reverb Modelling layout: home --- -

WORK IN PROGRESS

+## Abstract +Reverberation is a key element in spatial audio perception, histor- ically achieved with the use of analogue devices, such as plate and spring reverb, and in the last decades with digital signal process- ing techniques that have allowed different approaches for Virtual Analogue Modelling (VAM). The electromechanical functioning of the spring reverb makes it a nonlinear system that is difficult to fully emulate in the digital domain with white-box modelling techniques. In this study, we compare five different neural net- work architectures, including convolutional and recurrent models, to assess their effectiveness in replicating the characteristics of this audio effect. The evaluation is conducted on two datasets at sam- pling rates of 16 kHz and 48 kHz. This paper specifically focuses on neural audio architectures that offer parametric control, aiming to advance the boundaries of current black-box modelling tech- niques in the domain of spring reverberation. -## Audio Examples compared with target data +## Audio Examples collected during the evaluation -| Target | GCN | TCN | WaveNet | GRU | LSTM | -|--------|-----|-----|---------|-----|------| -| | | | | | | +| Model | Target | Prediction | +|-------|--------|------------| +| GCN | | | +| TCN | | | +| WaveNet | | | +| GRU | | | +| LSTM | | | -## Audio Examples with other kind of sounds +## Audio Examples with other kind of sounds obtained at inference | Sound | Dry | GCN | TCN | WaveNet | GRU | LSTM | |-------|-----|-----|-----|---------|-----|------| -| Drums | | | | | | | -| Synth | | | | | | | +| Drums | | | | | | | +| Synth | | | | | | |

Bibtex Citation