-
-
Notifications
You must be signed in to change notification settings - Fork 766
Commit
- Loading branch information
There are no files selected for viewing
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,44 @@ | ||
# MIT License | ||
# | ||
# Copyright (c) 2020 CNRS | ||
# | ||
# Permission is hereby granted, free of charge, to any person obtaining a copy | ||
# of this software and associated documentation files (the "Software"), to deal | ||
# in the Software without restriction, including without limitation the rights | ||
# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell | ||
# copies of the Software, and to permit persons to whom the Software is | ||
# furnished to do so, subject to the following conditions: | ||
# | ||
# The above copyright notice and this permission notice shall be included in all | ||
# copies or substantial portions of the Software. | ||
# | ||
# THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR | ||
# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, | ||
# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE | ||
# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER | ||
# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, | ||
# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE | ||
# SOFTWARE. | ||
|
||
|
||
from typing import Optional | ||
|
||
import torch | ||
import torch.nn as nn | ||
import torch.nn.functional as F | ||
from transformers import AutoModel | ||
This comment has been minimized.
Sorry, something went wrong.
This comment has been minimized.
Sorry, something went wrong.
SevKod
Author
Contributor
|
||
|
||
class WavLM(nn.Module): | ||
|
||
def __init__(self): | ||
super().__init__() | ||
|
||
self.wvlm = AutoModel.from_pretrained('microsoft/wavlm-base') #Load the model | ||
|
||
def forward(self, waveforms: torch.Tensor) -> torch.Tensor: | ||
|
||
waveforms = torch.squeeze(waveforms,1) #waveforms : (batch, channel, sample) -> (batch,sample) | ||
with torch.no_grad(): | ||
outputs = self.wvlm(waveforms).extract_features #Compute the features and extract last hidden layer weights | ||
|
||
return (outputs) |
Did you also try the model from
torchaudio
?That would remove this additional
transformers
dependency.Other criteria to decide between
torchaudio
andtransformers
include: