From 8428f9e58b9fd99f62fae6790fdae34909081268 Mon Sep 17 00:00:00 2001 From: khaled Date: Wed, 9 Mar 2022 18:08:45 +0100 Subject: [PATCH] update to 0.0.16 --- README.md | 2 +- hear21passt/__init__.py | 2 +- hear21passt/hop160base.py | 2 +- hear21passt/hop160base2lvl.py | 2 +- hear21passt/hop160base2lvlmel.py | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) diff --git a/README.md b/README.md index c1eba56..bfea4e2 100644 --- a/README.md +++ b/README.md @@ -13,7 +13,7 @@ pip3 install torch==1.8.1+cu111 torchaudio==0.8.1 -f https://download.pytorch.o # Installation Install the latest version of this repo: ```shell -pip install -e 'git+https://github.com/kkoutini/passt_hear21@0.0.15#egg=hear21passt' +pip install -e 'git+https://github.com/kkoutini/passt_hear21@0.0.16#egg=hear21passt' ``` The models follow the [common API](https://neuralaudio.ai/hear2021-holistic-evaluation-of-audio-representations.html#common-api) of HEAR 21 diff --git a/hear21passt/__init__.py b/hear21passt/__init__.py index d05ffc1..1e13e57 100644 --- a/hear21passt/__init__.py +++ b/hear21passt/__init__.py @@ -1,5 +1,5 @@ -__version__ = "0.0.15" +__version__ = "0.0.16" def embeding_size(hop=50, embeding_size=1000): diff --git a/hear21passt/hop160base.py b/hear21passt/hop160base.py index 55015c5..410830c 100644 --- a/hear21passt/hop160base.py +++ b/hear21passt/hop160base.py @@ -49,7 +49,7 @@ def get_basic_timestamp_embeddings(audio, model): def get_basic_model(**kwargs): - mel = AugmentMelSTFT(n_mels=128, sr=32000, win_length=800, hopsize=320, n_fft=1024, freqm=48, + mel = AugmentMelSTFT(n_mels=128, sr=32000, win_length=800, hopsize=160, n_fft=1024, freqm=48, timem=192, htk=False, fmin=0.0, fmax=None, norm=1, fmin_aug_range=10, fmax_aug_range=2000) diff --git a/hear21passt/hop160base2lvl.py b/hear21passt/hop160base2lvl.py index a981e5f..f0556e0 100644 --- a/hear21passt/hop160base2lvl.py +++ b/hear21passt/hop160base2lvl.py @@ -41,7 +41,7 @@ def get_timestamp_embeddings(audio, model): def get_2lvl_model(**kwargs): - mel = AugmentMelSTFT(n_mels=128, sr=32000, win_length=800, hopsize=100, n_fft=1024, freqm=48, + mel = AugmentMelSTFT(n_mels=128, sr=32000, win_length=800, hopsize=160, n_fft=1024, freqm=48, timem=192, htk=False, fmin=0.0, fmax=None, norm=1, fmin_aug_range=10, fmax_aug_range=2000) diff --git a/hear21passt/hop160base2lvlmel.py b/hear21passt/hop160base2lvlmel.py index da9b7a0..cc0daf6 100644 --- a/hear21passt/hop160base2lvlmel.py +++ b/hear21passt/hop160base2lvlmel.py @@ -42,7 +42,7 @@ def get_timestamp_embeddings(audio, model): def get_2lvl_model(**kwargs): - mel = AugmentMelSTFT(n_mels=128, sr=32000, win_length=800, hopsize=100, n_fft=1024, freqm=48, + mel = AugmentMelSTFT(n_mels=128, sr=32000, win_length=800, hopsize=160, n_fft=1024, freqm=48, timem=192, htk=False, fmin=0.0, fmax=None, norm=1, fmin_aug_range=10, fmax_aug_range=2000)