vggish
copied
You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Readme
Files and versions
65 lines
2.0 KiB
65 lines
2.0 KiB
# Copyright 2021 Zilliz. All rights reserved.
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
import logging
|
|
import os
|
|
import numpy
|
|
from pathlib import Path
|
|
|
|
import torch
|
|
|
|
from towhee.operator.base import NNOperator
|
|
from towhee.models.vggish.torch_vggish import VGG
|
|
from towhee import register
|
|
|
|
import vggish_input
|
|
|
|
import warnings
|
|
|
|
warnings.filterwarnings('ignore')
|
|
log = logging.getLogger()
|
|
|
|
|
|
@register(output_schema=['vec'])
|
|
class Vggish(NNOperator):
|
|
"""
|
|
"""
|
|
def __init__(self, weights_path: str = None, framework: str = 'pytorch') -> None:
|
|
super().__init__(framework=framework)
|
|
self.device = "cuda" if torch.cuda.is_available() else "cpu"
|
|
self.model = VGG()
|
|
if not weights_path:
|
|
path = str(Path(__file__).parent)
|
|
weights_path = os.path.join(path, 'vggish.pth')
|
|
state_dict = torch.load(weights_path, map_location=torch.device('cpu'))
|
|
self.model.load_state_dict(state_dict)
|
|
self.model.eval()
|
|
self.model.to(self.device)
|
|
|
|
def __call__(self, audio: str) -> numpy.ndarray:
|
|
audio_tensors = self.preprocess(audio).to(self.device)
|
|
features = self.model(audio_tensors)
|
|
outs = features.to("cpu")
|
|
return outs.detach().numpy()
|
|
|
|
def preprocess(self, audio_path: str):
|
|
audio_tensors = vggish_input.wavfile_to_examples(audio_path)
|
|
return audio_tensors
|
|
|
|
|
|
# if __name__ == '__main__':
|
|
# encoder = Vggish()
|
|
# audio_path = '/path/to/audio/wav'
|
|
# vec = encoder(audio_path)
|
|
# print(vec.shape)
|