diff --git a/configs/model/audio_encoder/htsat.yaml b/configs/model/audio_encoder/htsat.yaml index 4c9eab1..bbc53be 100644 --- a/configs/model/audio_encoder/htsat.yaml +++ b/configs/model/audio_encoder/htsat.yaml @@ -7,7 +7,7 @@ encoder: mel_bins: 64 spec_window_size: 1024 hop_size: 480 - patch_size: 16 + patch_size: 4 embed_dim: 128 depths: - 2