☢️ Audiomer ☢️
Audiomer: A Convolutional Transformer for Keyword Spotting
[ arXiv ] |
[ Previous SOTA ] |
[ Model Architecture ] |
|---|
NOTE: This is a pre-print release, the code might have bugs.
Results on SpeechCommands
Model Architecture
Performer Conv-Attention
Usage
To reproduce the results in the paper, follow the instructions:
- To download the Speech Commands v2 dataset, run:
python3 datamodules/SpeechCommands12.py - To train Audiomer-S and Audiomer-L on all three datasets thrice, run:
python3 run_expts.py - To evaluate a model on a dataset, run:
python3 evaluate.py --checkpoint_path /path/to/checkpoint.ckpt --model <model type> --dataset <name of dataset>. - For example:
python3 evaluate.py --checkpoint_path ./epoch=300.ckpt --model S --dataset SC20
System requirements
- NVIDIA GPU with CUDA
- Python 3.6 or higher.
- pytorch_lightning
- torchaudio
- performer_pytorch