pipelines.rst 2.82 KB
Newer Older
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
torchaudio.pipelines
====================

.. currentmodule:: torchaudio.pipelines

The pipelines subpackage contains API to access the models with pretrained weights, and information/helper functions associated the pretrained weights.

wav2vec 2.0 / HuBERT - Representation Learning
----------------------------------------------

.. autoclass:: Wav2Vec2Bundle

   .. automethod:: get_model

WAV2VEC2_BASE
16
~~~~~~~~~~~~~
17
18
19
20
21
22
23

.. container:: py attribute

   .. autodata:: WAV2VEC2_BASE
      :no-value:

WAV2VEC2_LARGE
24
~~~~~~~~~~~~~~
25
26
27
28
29
30
31

.. container:: py attribute

   .. autodata:: WAV2VEC2_LARGE
      :no-value:

WAV2VEC2_LARGE_LV60K
32
~~~~~~~~~~~~~~~~~~~~
33
34
35
36
37
38
39
40

.. container:: py attribute

   .. autodata:: WAV2VEC2_LARGE_LV60K
      :no-value:


WAV2VEC2_XLSR53
41
~~~~~~~~~~~~~~~
42
43
44
45
46
47
48

.. container:: py attribute

   .. autodata:: WAV2VEC2_XLSR53
      :no-value:

HUBERT_BASE
49
~~~~~~~~~~~
50
51
52
53
54
55
56

.. container:: py attribute

   .. autodata:: HUBERT_BASE
      :no-value:

HUBERT_LARGE
57
~~~~~~~~~~~~
58
59
60
61
62
63
64

.. container:: py attribute

   .. autodata:: HUBERT_LARGE
      :no-value:

HUBERT_XLARGE
65
~~~~~~~~~~~~~
66
67
68
69
70
71

.. container:: py attribute

   .. autodata:: HUBERT_XLARGE
      :no-value:

72
73
wav2vec 2.0 / HuBERT - Fine-tuned ASR
-------------------------------------
74
75
76
77
78
79
80
81
82

.. autoclass:: Wav2Vec2ASRBundle

   .. automethod:: get_model

   .. automethod:: get_labels


WAV2VEC2_ASR_BASE_10M
83
~~~~~~~~~~~~~~~~~~~~~
84
85
86
87
88
89
90

.. container:: py attribute

   .. autodata:: WAV2VEC2_ASR_BASE_10M
      :no-value:

WAV2VEC2_ASR_BASE_100H
91
~~~~~~~~~~~~~~~~~~~~~~
92
93
94
95
96
97
98
      
.. container:: py attribute

   .. autodata:: WAV2VEC2_ASR_BASE_100H
      :no-value:

WAV2VEC2_ASR_BASE_960H
99
~~~~~~~~~~~~~~~~~~~~~~
100
101
102
103
104
105
106

.. container:: py attribute

   .. autodata:: WAV2VEC2_ASR_BASE_960H
      :no-value:

WAV2VEC2_ASR_LARGE_10M
107
~~~~~~~~~~~~~~~~~~~~~~
108
109
110
111
112
113
114

.. container:: py attribute

   .. autodata:: WAV2VEC2_ASR_LARGE_10M
      :no-value:

WAV2VEC2_ASR_LARGE_100H
115
~~~~~~~~~~~~~~~~~~~~~~~
116
117
118
119
120
121
122

.. container:: py attribute

   .. autodata:: WAV2VEC2_ASR_LARGE_100H
      :no-value:

WAV2VEC2_ASR_LARGE_960H
123
~~~~~~~~~~~~~~~~~~~~~~~
124
125
126
127
128
129
130

.. container:: py attribute

   .. autodata:: WAV2VEC2_ASR_LARGE_960H
      :no-value:

WAV2VEC2_ASR_LARGE_LV60K_10M
131
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
132
133
134
135
136
137
138

.. container:: py attribute

   .. autodata:: WAV2VEC2_ASR_LARGE_LV60K_10M
      :no-value:

WAV2VEC2_ASR_LARGE_LV60K_100H
139
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
140
141
142
143
144
145
146

.. container:: py attribute

   .. autodata:: WAV2VEC2_ASR_LARGE_LV60K_100H
      :no-value:

WAV2VEC2_ASR_LARGE_LV60K_960H
147
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
148
149
150
151
152
153
154

.. container:: py attribute

   .. autodata:: WAV2VEC2_ASR_LARGE_LV60K_960H
      :no-value:

HUBERT_ASR_LARGE
155
~~~~~~~~~~~~~~~~
156
157
158
159
160
161
162

.. container:: py attribute

   .. autodata:: HUBERT_ASR_LARGE
      :no-value:

HUBERT_ASR_XLARGE
163
~~~~~~~~~~~~~~~~~
164
165
166
167
168
169
170
171
172
173

.. container:: py attribute

   .. autodata:: HUBERT_ASR_XLARGE
      :no-value:

References
----------

.. footbibliography::