-
Notifications
You must be signed in to change notification settings - Fork 126
/
audio_classifier.py
133 lines (107 loc) · 4.75 KB
/
audio_classifier.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
# Copyright 2022 The TensorFlow Authors. 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.
"""Audio classifier task."""
import dataclasses
from tensorflow_lite_support.python.task.audio.core import audio_record
from tensorflow_lite_support.python.task.audio.core import tensor_audio
from tensorflow_lite_support.python.task.audio.core.pybinds import _pywrap_audio_buffer
from tensorflow_lite_support.python.task.audio.pybinds import _pywrap_audio_classifier
from tensorflow_lite_support.python.task.core.proto import base_options_pb2
from tensorflow_lite_support.python.task.processor.proto import classification_options_pb2
from tensorflow_lite_support.python.task.processor.proto import classifications_pb2
_CppAudioFormat = _pywrap_audio_buffer.AudioFormat
_CppAudioBuffer = _pywrap_audio_buffer.AudioBuffer
_CppAudioClassifier = _pywrap_audio_classifier.AudioClassifier
_ClassificationOptions = classification_options_pb2.ClassificationOptions
_BaseOptions = base_options_pb2.BaseOptions
@dataclasses.dataclass
class AudioClassifierOptions:
"""Options for the audio classifier task."""
base_options: _BaseOptions
classification_options: _ClassificationOptions = _ClassificationOptions()
class AudioClassifier(object):
"""Class that performs classification on audio."""
def __init__(self, options: AudioClassifierOptions,
classifier: _CppAudioClassifier) -> None:
"""Initializes the `AudioClassifier` object."""
# Creates the object of C++ AudioClassifier class.
self._options = options
self._classifier = classifier
@classmethod
def create_from_file(cls, file_path: str) -> "AudioClassifier":
"""Creates the `AudioClassifier` object from a TensorFlow Lite model.
Args:
file_path: Path to the model.
Returns:
`AudioClassifier` object that's created from `options`.
Raises:
RuntimeError if failed to create `AudioClassifier` object from the
provided file such as invalid file.
"""
base_options = _BaseOptions(file_name=file_path)
options = AudioClassifierOptions(base_options=base_options)
return cls.create_from_options(options)
@classmethod
def create_from_options(cls,
options: AudioClassifierOptions) -> "AudioClassifier":
"""Creates the `AudioClassifier` object from audio classifier options.
Args:
options: Options for the audio classifier task.
Returns:
`AudioClassifier` object that's created from `options`.
Raises:
RuntimeError if failed to create `AudioClassifier` object from
`AudioClassifierOptions` such as missing the model.
"""
classifier = _CppAudioClassifier.create_from_options(
options.base_options, options.classification_options)
return cls(options, classifier)
def create_input_tensor_audio(self) -> tensor_audio.TensorAudio:
"""Creates a TensorAudio instance to store the audio input.
Returns:
A TensorAudio instance.
"""
return tensor_audio.TensorAudio(
audio_format=self.required_audio_format,
buffer_size=self.required_input_buffer_size)
def create_audio_record(self) -> audio_record.AudioRecord:
"""Creates an AudioRecord instance to record audio.
Returns:
An AudioRecord instance.
"""
return audio_record.AudioRecord(self.required_audio_format.channels,
self.required_audio_format.sample_rate,
self.required_input_buffer_size)
def classify(
self,
audio: tensor_audio.TensorAudio,
) -> classifications_pb2.ClassificationResult:
"""Performs classification on the provided TensorAudio.
Args:
audio: Tensor audio, used to extract the feature vectors.
Returns:
classification result.
Raises:
RuntimeError if failed to get the feature vector.
"""
return self._classifier.classify(
_CppAudioBuffer(audio.buffer, audio.buffer_size, audio.format))
@property
def required_input_buffer_size(self) -> int:
"""Gets the required input buffer size for the model."""
return self._classifier.get_required_input_buffer_size()
@property
def required_audio_format(self) -> _CppAudioFormat:
"""Gets the required audio format for the model."""
return self._classifier.get_required_audio_format()