simulstream 0.1.0__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- docs/source/conf.py +47 -0
- simulstream/__init__.py +15 -0
- simulstream/client/__init__.py +0 -0
- simulstream/client/wav_reader_client.py +228 -0
- simulstream/config.py +31 -0
- simulstream/inference.py +170 -0
- simulstream/metrics/__init__.py +0 -0
- simulstream/metrics/detokenizers.py +71 -0
- simulstream/metrics/logger.py +32 -0
- simulstream/metrics/readers.py +348 -0
- simulstream/metrics/score_latency.py +130 -0
- simulstream/metrics/score_quality.py +169 -0
- simulstream/metrics/scorers/__init__.py +0 -0
- simulstream/metrics/scorers/latency/__init__.py +115 -0
- simulstream/metrics/scorers/latency/mwersegmenter.py +136 -0
- simulstream/metrics/scorers/latency/stream_laal.py +119 -0
- simulstream/metrics/scorers/quality/__init__.py +132 -0
- simulstream/metrics/scorers/quality/comet.py +57 -0
- simulstream/metrics/scorers/quality/mwersegmenter.py +93 -0
- simulstream/metrics/scorers/quality/sacrebleu.py +59 -0
- simulstream/metrics/stats.py +184 -0
- simulstream/server/__init__.py +0 -0
- simulstream/server/http_server.py +95 -0
- simulstream/server/message_processor.py +156 -0
- simulstream/server/speech_processors/__init__.py +173 -0
- simulstream/server/speech_processors/base.py +135 -0
- simulstream/server/speech_processors/base_streamatt.py +320 -0
- simulstream/server/speech_processors/canary_sliding_window_retranslation.py +73 -0
- simulstream/server/speech_processors/hf_sliding_window_retranslation.py +87 -0
- simulstream/server/speech_processors/incremental_output.py +85 -0
- simulstream/server/speech_processors/seamless_sliding_window_retranslation.py +84 -0
- simulstream/server/speech_processors/seamless_streamatt.py +268 -0
- simulstream/server/speech_processors/simuleval_wrapper.py +165 -0
- simulstream/server/speech_processors/sliding_window_retranslation.py +135 -0
- simulstream/server/speech_processors/vad_wrapper.py +180 -0
- simulstream/server/websocket_server.py +236 -0
- simulstream-0.1.0.dist-info/METADATA +465 -0
- simulstream-0.1.0.dist-info/RECORD +48 -0
- simulstream-0.1.0.dist-info/WHEEL +5 -0
- simulstream-0.1.0.dist-info/entry_points.txt +8 -0
- simulstream-0.1.0.dist-info/licenses/LICENSE +201 -0
- simulstream-0.1.0.dist-info/top_level.txt +3 -0
- uts/__init__.py +0 -0
- uts/metrics/__init__.py +0 -0
- uts/metrics/log_reader.py +50 -0
- uts/speech_processors/__init__.py +0 -0
- uts/speech_processors/test_simuleval_wrapper.py +88 -0
- uts/utils.py +5 -0
|
@@ -0,0 +1,465 @@
|
|
|
1
|
+
Metadata-Version: 2.4
|
|
2
|
+
Name: simulstream
|
|
3
|
+
Version: 0.1.0
|
|
4
|
+
Summary: A server to run simultaneous/streaming experiments and demo
|
|
5
|
+
Author-email: Marco Gaido <mgaido@fbk.eu>, FBK HLT-MT <mt@fbk.eu>
|
|
6
|
+
License: Apache License
|
|
7
|
+
Version 2.0, January 2004
|
|
8
|
+
http://www.apache.org/licenses/
|
|
9
|
+
|
|
10
|
+
TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
|
|
11
|
+
|
|
12
|
+
1. Definitions.
|
|
13
|
+
|
|
14
|
+
"License" shall mean the terms and conditions for use, reproduction,
|
|
15
|
+
and distribution as defined by Sections 1 through 9 of this document.
|
|
16
|
+
|
|
17
|
+
"Licensor" shall mean the copyright owner or entity authorized by
|
|
18
|
+
the copyright owner that is granting the License.
|
|
19
|
+
|
|
20
|
+
"Legal Entity" shall mean the union of the acting entity and all
|
|
21
|
+
other entities that control, are controlled by, or are under common
|
|
22
|
+
control with that entity. For the purposes of this definition,
|
|
23
|
+
"control" means (i) the power, direct or indirect, to cause the
|
|
24
|
+
direction or management of such entity, whether by contract or
|
|
25
|
+
otherwise, or (ii) ownership of fifty percent (50%) or more of the
|
|
26
|
+
outstanding shares, or (iii) beneficial ownership of such entity.
|
|
27
|
+
|
|
28
|
+
"You" (or "Your") shall mean an individual or Legal Entity
|
|
29
|
+
exercising permissions granted by this License.
|
|
30
|
+
|
|
31
|
+
"Source" form shall mean the preferred form for making modifications,
|
|
32
|
+
including but not limited to software source code, documentation
|
|
33
|
+
source, and configuration files.
|
|
34
|
+
|
|
35
|
+
"Object" form shall mean any form resulting from mechanical
|
|
36
|
+
transformation or translation of a Source form, including but
|
|
37
|
+
not limited to compiled object code, generated documentation,
|
|
38
|
+
and conversions to other media types.
|
|
39
|
+
|
|
40
|
+
"Work" shall mean the work of authorship, whether in Source or
|
|
41
|
+
Object form, made available under the License, as indicated by a
|
|
42
|
+
copyright notice that is included in or attached to the work
|
|
43
|
+
(an example is provided in the Appendix below).
|
|
44
|
+
|
|
45
|
+
"Derivative Works" shall mean any work, whether in Source or Object
|
|
46
|
+
form, that is based on (or derived from) the Work and for which the
|
|
47
|
+
editorial revisions, annotations, elaborations, or other modifications
|
|
48
|
+
represent, as a whole, an original work of authorship. For the purposes
|
|
49
|
+
of this License, Derivative Works shall not include works that remain
|
|
50
|
+
separable from, or merely link (or bind by name) to the interfaces of,
|
|
51
|
+
the Work and Derivative Works thereof.
|
|
52
|
+
|
|
53
|
+
"Contribution" shall mean any work of authorship, including
|
|
54
|
+
the original version of the Work and any modifications or additions
|
|
55
|
+
to that Work or Derivative Works thereof, that is intentionally
|
|
56
|
+
submitted to Licensor for inclusion in the Work by the copyright owner
|
|
57
|
+
or by an individual or Legal Entity authorized to submit on behalf of
|
|
58
|
+
the copyright owner. For the purposes of this definition, "submitted"
|
|
59
|
+
means any form of electronic, verbal, or written communication sent
|
|
60
|
+
to the Licensor or its representatives, including but not limited to
|
|
61
|
+
communication on electronic mailing lists, source code control systems,
|
|
62
|
+
and issue tracking systems that are managed by, or on behalf of, the
|
|
63
|
+
Licensor for the purpose of discussing and improving the Work, but
|
|
64
|
+
excluding communication that is conspicuously marked or otherwise
|
|
65
|
+
designated in writing by the copyright owner as "Not a Contribution."
|
|
66
|
+
|
|
67
|
+
"Contributor" shall mean Licensor and any individual or Legal Entity
|
|
68
|
+
on behalf of whom a Contribution has been received by Licensor and
|
|
69
|
+
subsequently incorporated within the Work.
|
|
70
|
+
|
|
71
|
+
2. Grant of Copyright License. Subject to the terms and conditions of
|
|
72
|
+
this License, each Contributor hereby grants to You a perpetual,
|
|
73
|
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
|
74
|
+
copyright license to reproduce, prepare Derivative Works of,
|
|
75
|
+
publicly display, publicly perform, sublicense, and distribute the
|
|
76
|
+
Work and such Derivative Works in Source or Object form.
|
|
77
|
+
|
|
78
|
+
3. Grant of Patent License. Subject to the terms and conditions of
|
|
79
|
+
this License, each Contributor hereby grants to You a perpetual,
|
|
80
|
+
worldwide, non-exclusive, no-charge, royalty-free, irrevocable
|
|
81
|
+
(except as stated in this section) patent license to make, have made,
|
|
82
|
+
use, offer to sell, sell, import, and otherwise transfer the Work,
|
|
83
|
+
where such license applies only to those patent claims licensable
|
|
84
|
+
by such Contributor that are necessarily infringed by their
|
|
85
|
+
Contribution(s) alone or by combination of their Contribution(s)
|
|
86
|
+
with the Work to which such Contribution(s) was submitted. If You
|
|
87
|
+
institute patent litigation against any entity (including a
|
|
88
|
+
cross-claim or counterclaim in a lawsuit) alleging that the Work
|
|
89
|
+
or a Contribution incorporated within the Work constitutes direct
|
|
90
|
+
or contributory patent infringement, then any patent licenses
|
|
91
|
+
granted to You under this License for that Work shall terminate
|
|
92
|
+
as of the date such litigation is filed.
|
|
93
|
+
|
|
94
|
+
4. Redistribution. You may reproduce and distribute copies of the
|
|
95
|
+
Work or Derivative Works thereof in any medium, with or without
|
|
96
|
+
modifications, and in Source or Object form, provided that You
|
|
97
|
+
meet the following conditions:
|
|
98
|
+
|
|
99
|
+
(a) You must give any other recipients of the Work or
|
|
100
|
+
Derivative Works a copy of this License; and
|
|
101
|
+
|
|
102
|
+
(b) You must cause any modified files to carry prominent notices
|
|
103
|
+
stating that You changed the files; and
|
|
104
|
+
|
|
105
|
+
(c) You must retain, in the Source form of any Derivative Works
|
|
106
|
+
that You distribute, all copyright, patent, trademark, and
|
|
107
|
+
attribution notices from the Source form of the Work,
|
|
108
|
+
excluding those notices that do not pertain to any part of
|
|
109
|
+
the Derivative Works; and
|
|
110
|
+
|
|
111
|
+
(d) If the Work includes a "NOTICE" text file as part of its
|
|
112
|
+
distribution, then any Derivative Works that You distribute must
|
|
113
|
+
include a readable copy of the attribution notices contained
|
|
114
|
+
within such NOTICE file, excluding those notices that do not
|
|
115
|
+
pertain to any part of the Derivative Works, in at least one
|
|
116
|
+
of the following places: within a NOTICE text file distributed
|
|
117
|
+
as part of the Derivative Works; within the Source form or
|
|
118
|
+
documentation, if provided along with the Derivative Works; or,
|
|
119
|
+
within a display generated by the Derivative Works, if and
|
|
120
|
+
wherever such third-party notices normally appear. The contents
|
|
121
|
+
of the NOTICE file are for informational purposes only and
|
|
122
|
+
do not modify the License. You may add Your own attribution
|
|
123
|
+
notices within Derivative Works that You distribute, alongside
|
|
124
|
+
or as an addendum to the NOTICE text from the Work, provided
|
|
125
|
+
that such additional attribution notices cannot be construed
|
|
126
|
+
as modifying the License.
|
|
127
|
+
|
|
128
|
+
You may add Your own copyright statement to Your modifications and
|
|
129
|
+
may provide additional or different license terms and conditions
|
|
130
|
+
for use, reproduction, or distribution of Your modifications, or
|
|
131
|
+
for any such Derivative Works as a whole, provided Your use,
|
|
132
|
+
reproduction, and distribution of the Work otherwise complies with
|
|
133
|
+
the conditions stated in this License.
|
|
134
|
+
|
|
135
|
+
5. Submission of Contributions. Unless You explicitly state otherwise,
|
|
136
|
+
any Contribution intentionally submitted for inclusion in the Work
|
|
137
|
+
by You to the Licensor shall be under the terms and conditions of
|
|
138
|
+
this License, without any additional terms or conditions.
|
|
139
|
+
Notwithstanding the above, nothing herein shall supersede or modify
|
|
140
|
+
the terms of any separate license agreement you may have executed
|
|
141
|
+
with Licensor regarding such Contributions.
|
|
142
|
+
|
|
143
|
+
6. Trademarks. This License does not grant permission to use the trade
|
|
144
|
+
names, trademarks, service marks, or product names of the Licensor,
|
|
145
|
+
except as required for reasonable and customary use in describing the
|
|
146
|
+
origin of the Work and reproducing the content of the NOTICE file.
|
|
147
|
+
|
|
148
|
+
7. Disclaimer of Warranty. Unless required by applicable law or
|
|
149
|
+
agreed to in writing, Licensor provides the Work (and each
|
|
150
|
+
Contributor provides its Contributions) on an "AS IS" BASIS,
|
|
151
|
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
|
|
152
|
+
implied, including, without limitation, any warranties or conditions
|
|
153
|
+
of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
|
|
154
|
+
PARTICULAR PURPOSE. You are solely responsible for determining the
|
|
155
|
+
appropriateness of using or redistributing the Work and assume any
|
|
156
|
+
risks associated with Your exercise of permissions under this License.
|
|
157
|
+
|
|
158
|
+
8. Limitation of Liability. In no event and under no legal theory,
|
|
159
|
+
whether in tort (including negligence), contract, or otherwise,
|
|
160
|
+
unless required by applicable law (such as deliberate and grossly
|
|
161
|
+
negligent acts) or agreed to in writing, shall any Contributor be
|
|
162
|
+
liable to You for damages, including any direct, indirect, special,
|
|
163
|
+
incidental, or consequential damages of any character arising as a
|
|
164
|
+
result of this License or out of the use or inability to use the
|
|
165
|
+
Work (including but not limited to damages for loss of goodwill,
|
|
166
|
+
work stoppage, computer failure or malfunction, or any and all
|
|
167
|
+
other commercial damages or losses), even if such Contributor
|
|
168
|
+
has been advised of the possibility of such damages.
|
|
169
|
+
|
|
170
|
+
9. Accepting Warranty or Additional Liability. While redistributing
|
|
171
|
+
the Work or Derivative Works thereof, You may choose to offer,
|
|
172
|
+
and charge a fee for, acceptance of support, warranty, indemnity,
|
|
173
|
+
or other liability obligations and/or rights consistent with this
|
|
174
|
+
License. However, in accepting such obligations, You may act only
|
|
175
|
+
on Your own behalf and on Your sole responsibility, not on behalf
|
|
176
|
+
of any other Contributor, and only if You agree to indemnify,
|
|
177
|
+
defend, and hold each Contributor harmless for any liability
|
|
178
|
+
incurred by, or claims asserted against, such Contributor by reason
|
|
179
|
+
of your accepting any such warranty or additional liability.
|
|
180
|
+
|
|
181
|
+
END OF TERMS AND CONDITIONS
|
|
182
|
+
|
|
183
|
+
APPENDIX: How to apply the Apache License to your work.
|
|
184
|
+
|
|
185
|
+
To apply the Apache License to your work, attach the following
|
|
186
|
+
boilerplate notice, with the fields enclosed by brackets "[]"
|
|
187
|
+
replaced with your own identifying information. (Don't include
|
|
188
|
+
the brackets!) The text should be enclosed in the appropriate
|
|
189
|
+
comment syntax for the file format. We also recommend that a
|
|
190
|
+
file or class name and description of purpose be included on the
|
|
191
|
+
same "printed page" as the copyright notice for easier
|
|
192
|
+
identification within third-party archives.
|
|
193
|
+
|
|
194
|
+
Copyright 2023 FBK
|
|
195
|
+
|
|
196
|
+
Licensed under the Apache License, Version 2.0 (the "License");
|
|
197
|
+
you may not use this file except in compliance with the License.
|
|
198
|
+
You may obtain a copy of the License at
|
|
199
|
+
|
|
200
|
+
http://www.apache.org/licenses/LICENSE-2.0
|
|
201
|
+
|
|
202
|
+
Unless required by applicable law or agreed to in writing, software
|
|
203
|
+
distributed under the License is distributed on an "AS IS" BASIS,
|
|
204
|
+
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
205
|
+
See the License for the specific language governing permissions and
|
|
206
|
+
limitations under the License.
|
|
207
|
+
|
|
208
|
+
Project-URL: GitHub, https://github.com/hlt-mt/simulstream
|
|
209
|
+
Project-URL: Issues, https://github.com/hlt-mt/simulstream/issues
|
|
210
|
+
Keywords: simultaneous,demo,websocket,streaming
|
|
211
|
+
Requires-Python: >=3.9
|
|
212
|
+
Description-Content-Type: text/markdown
|
|
213
|
+
License-File: LICENSE
|
|
214
|
+
Requires-Dist: pyyaml>6.0
|
|
215
|
+
Requires-Dist: websockets
|
|
216
|
+
Requires-Dist: torch
|
|
217
|
+
Requires-Dist: librosa
|
|
218
|
+
Provides-Extra: dev
|
|
219
|
+
Requires-Dist: pytest==7.4.0; extra == "dev"
|
|
220
|
+
Requires-Dist: flake8; extra == "dev"
|
|
221
|
+
Requires-Dist: sphinx; extra == "dev"
|
|
222
|
+
Requires-Dist: sphinx-rtd-theme; extra == "dev"
|
|
223
|
+
Requires-Dist: build; extra == "dev"
|
|
224
|
+
Requires-Dist: twine; extra == "dev"
|
|
225
|
+
Provides-Extra: hf
|
|
226
|
+
Requires-Dist: transformers==4.48.1; extra == "hf"
|
|
227
|
+
Provides-Extra: canary
|
|
228
|
+
Requires-Dist: Cython; extra == "canary"
|
|
229
|
+
Requires-Dist: nemo_toolkit[asr]==2.4.0; extra == "canary"
|
|
230
|
+
Provides-Extra: vad
|
|
231
|
+
Requires-Dist: silero-vad; extra == "vad"
|
|
232
|
+
Provides-Extra: eval
|
|
233
|
+
Requires-Dist: unbabel-comet==2.2.6; extra == "eval"
|
|
234
|
+
Requires-Dist: mweralign; extra == "eval"
|
|
235
|
+
Requires-Dist: sacrebleu; extra == "eval"
|
|
236
|
+
Dynamic: license-file
|
|
237
|
+
|
|
238
|
+
<div align="center">
|
|
239
|
+
<img src="https://github.com/hlt-mt/simulstream/blob/main/docs/source/_static/logo.png?raw=true" width="200" alt="simulstream logo">
|
|
240
|
+
</div>
|
|
241
|
+
|
|
242
|
+
|
|
243
|
+
# simulstream
|
|
244
|
+
|
|
245
|
+
``simulstream`` is a Python library for simultaneous/streaming speech recognition and translation.
|
|
246
|
+
It enables both the simulation with existing files to score systems, like in the
|
|
247
|
+
[SimulEval](https://github.com/facebookresearch/SimulEval) project, and the possibility to run
|
|
248
|
+
demos on a browser.
|
|
249
|
+
|
|
250
|
+
``simulstream`` provides a WebSocket server and utilities for running streaming speech processing
|
|
251
|
+
experiments and demos. It supports real-time transcription and translation through streaming audio
|
|
252
|
+
input. By streaming, we mean that the library by default assumes that the input is an unbounded
|
|
253
|
+
speech signal, rather than many short speech segments as in simultaneous speech processing.
|
|
254
|
+
The simultaneous setting can be easily addressed by pre-segmenting the audio into many small
|
|
255
|
+
segments and feed each segment to ``simulstream``.
|
|
256
|
+
|
|
257
|
+
The repository is tested using Python 3.11. Although it may work also with other Python versions,
|
|
258
|
+
we do not ensure compatibility with them. Check out the [Usage](#Usage) section for instructions on
|
|
259
|
+
how to use the repository and the [Installation](#Installation) section for further information
|
|
260
|
+
about how to install the project.
|
|
261
|
+
|
|
262
|
+
|
|
263
|
+
## Installation
|
|
264
|
+
|
|
265
|
+
You can install the latest stable version from PyPI:
|
|
266
|
+
|
|
267
|
+
```shell
|
|
268
|
+
pip install simulstream
|
|
269
|
+
```
|
|
270
|
+
|
|
271
|
+
Or, to install from source:
|
|
272
|
+
|
|
273
|
+
```shell
|
|
274
|
+
git clone https://github.com/hlt-mt/simulstream.git
|
|
275
|
+
cd simulstream
|
|
276
|
+
pip install .
|
|
277
|
+
```
|
|
278
|
+
|
|
279
|
+
Please notice that these commands will only install the basic functionalities of the repository,
|
|
280
|
+
i.e. the WebSocket server and client. Additionally, you have to install the dependencies required
|
|
281
|
+
by the `speech processor` that you want to use. The repository comes with examples of speech
|
|
282
|
+
processors that rely on e.g. Transformers models or on the NVIDIA Canary model. You can install
|
|
283
|
+
the required dependencies for them by specifying the corresponding selector when installing
|
|
284
|
+
the repository. For instance, using Canary speech processors requires the ``canary`` selector.
|
|
285
|
+
But you can create your custom speech processor and install the corresponding dependencies.
|
|
286
|
+
|
|
287
|
+
Also the evaluation comes with additional dependencies, which can be installed with the ``eval``
|
|
288
|
+
selector. Be careful as the metrics include COMET, which has dependencies on Transformers and
|
|
289
|
+
other libraries that can run on conflict with those required by your speech processor.
|
|
290
|
+
|
|
291
|
+
As an example, if you want to install the ``simulstream`` package with Canary speech processors
|
|
292
|
+
and the evaluation package, run:
|
|
293
|
+
|
|
294
|
+
```shell
|
|
295
|
+
pip install -e .[canary,eval]
|
|
296
|
+
```
|
|
297
|
+
|
|
298
|
+
For development (with docs and testing tools):
|
|
299
|
+
|
|
300
|
+
```shell
|
|
301
|
+
pip install -e .[dev]
|
|
302
|
+
```
|
|
303
|
+
|
|
304
|
+
## Usage
|
|
305
|
+
|
|
306
|
+
The package is based on a WebSocket client-server interaction. The server waits for connections
|
|
307
|
+
by the clients. The clients can send configurations messages in JSON format (e.g. to specify
|
|
308
|
+
the input/output desired language) and audio chunks, encoded as 16-bit integers. The server
|
|
309
|
+
processes the input audio, generates the transcript/translation by means of the configured
|
|
310
|
+
speech processor and sends it to the client in JSON format. In addition, if configured to do so,
|
|
311
|
+
the server writes logs that can be used to compute metrics in a JSONL file.
|
|
312
|
+
|
|
313
|
+
To have everything up and running, first run the server and once the server is up and running
|
|
314
|
+
execute one or more clients. The repository contains the code for two types of clients: a web
|
|
315
|
+
interface client that can be used for demos and a python command-line client that can be used
|
|
316
|
+
for testing or evaluation of the speech processor performance.
|
|
317
|
+
|
|
318
|
+
Below, you can find a simple illustration of the overall architecture with the two scenarios:
|
|
319
|
+
|
|
320
|
+
<img src="docs/source/_static/architecture.png" alt="Overall simulstream architecture">
|
|
321
|
+
|
|
322
|
+
To streamline the evaluation of speech processors in terms of quality and latency, there is also
|
|
323
|
+
the possibility to run the inference with a configured speech processor without setting up a
|
|
324
|
+
client-server interaction. For more information, refer to the [Evaluation section](#Evaluation).
|
|
325
|
+
|
|
326
|
+
### WebSocket Server
|
|
327
|
+
|
|
328
|
+
Run the WebSocket server with YAML configuration files:
|
|
329
|
+
|
|
330
|
+
```shell
|
|
331
|
+
simulstream_server --server-config config/server.yaml \
|
|
332
|
+
--speech-processor-config config/speech_processor.yaml
|
|
333
|
+
```
|
|
334
|
+
|
|
335
|
+
The repository contains examples of YAML files both for the server and for some speech processors.
|
|
336
|
+
They can be edited and adapted.
|
|
337
|
+
|
|
338
|
+
#### Customize with Your Speech Processor
|
|
339
|
+
|
|
340
|
+
If you want to implement your own ``speech_processor``, you need to create a subclass
|
|
341
|
+
of ``simulstream.server.speech_processors.base.SpeechProcessor`` and add the class
|
|
342
|
+
to the ``PYTHONPATH``. Then reference the new class in the YAML file configuration
|
|
343
|
+
of the speech processor. Refer to the dedicated documentation for more
|
|
344
|
+
details on how to implement a speech processor in the project documentation and in the
|
|
345
|
+
docstring of both [``simulstream.server.speech_processors.base.SpeechProcessor``](simulstream/server/speech_processors/__init__.py)
|
|
346
|
+
and [`simulstream.server.speech_processors.base.BaseSpeechProcessor`](simulstream/server/speech_processors/base.py).
|
|
347
|
+
|
|
348
|
+
As reference, you can check the processors implemented in this repository, available in the
|
|
349
|
+
[`simulstream.server.speech_processors`](simulstream/server/speech_processors) module.
|
|
350
|
+
Notice that each speech processor can have additional dependencies that are not installed by
|
|
351
|
+
default with this codebase (see [Installation](#installation)).
|
|
352
|
+
|
|
353
|
+
|
|
354
|
+
### HTTP Server Web Client
|
|
355
|
+
|
|
356
|
+
For a demo, you can create an HTTP web server that servers a web interface interacting with the
|
|
357
|
+
WebSocket server. This can be done by:
|
|
358
|
+
|
|
359
|
+
```shell
|
|
360
|
+
simulstream_demo_http_server --config config/server.yaml -p 8001 --directory webdemo
|
|
361
|
+
```
|
|
362
|
+
|
|
363
|
+
In case you are not running the command from the root folder of this repository, you should change
|
|
364
|
+
accordingly the ``--directory`` parameter to point to the ``webdemo`` folder containing the HTML
|
|
365
|
+
files of the HTTP demo. You can of course replace ``8001`` with any other port number you prefer.
|
|
366
|
+
The web interface can then be accessed from the local laptop at ``http://localhost:8001/`` or from
|
|
367
|
+
any other terminal connected to the same network using the IP address of your workstation instead
|
|
368
|
+
of ``localhost``. Be careful not to use the same port specified for the WebSocket server if they
|
|
369
|
+
are running on the same machine. If running the HTTP server from a machine different from the one
|
|
370
|
+
where the WebSocket server runs, ensure that the HTTP server can connect to the WebSocket server
|
|
371
|
+
through the address specified in the ``config/server.yaml`` file.
|
|
372
|
+
|
|
373
|
+
|
|
374
|
+
### Python Client
|
|
375
|
+
|
|
376
|
+
|
|
377
|
+
If you want to process a set of audio files (e.g. a test set) with your system, send the audios
|
|
378
|
+
with the provided client:
|
|
379
|
+
|
|
380
|
+
```shell
|
|
381
|
+
simulstream_wavs_client --uri ws://localhost:8080/ \
|
|
382
|
+
--wav-list-file PATH_TO_TXT_FILE_WITH_A_LIST_OF_WAV_FILES.txt \
|
|
383
|
+
--tgt-lang it --src-lang en
|
|
384
|
+
```
|
|
385
|
+
|
|
386
|
+
Tne ``--uri`` should contain the address of the WebSocket client, so it should correspond to what
|
|
387
|
+
is specified in the server YAML config. The file specified in ``--wav-list-file`` should be a TXT
|
|
388
|
+
files containing for each line the path to a WAV file.
|
|
389
|
+
|
|
390
|
+
Before running the command, ensure that the logging of metrics is enabled in the configuration file
|
|
391
|
+
of the WebSocket server (i.e., ``metrics.enabled = True``) and that metrics are logged to an
|
|
392
|
+
empty/non-existing file, to ensure that the resulting file will contain only the logs related to
|
|
393
|
+
the files sent by the client. Then, compute the relevant scores as described below
|
|
394
|
+
([Evaluation](#Evaluation)).
|
|
395
|
+
|
|
396
|
+
|
|
397
|
+
### Evaluation
|
|
398
|
+
|
|
399
|
+
To evaluate a speech processor, as a first thing you need to run the inference on a test set.
|
|
400
|
+
This can be done either through the Python client (see section above) or through the dedicated
|
|
401
|
+
command that performs the inference without setting up a client-server interaction. This second
|
|
402
|
+
option can be followed by running the dedicated command:
|
|
403
|
+
|
|
404
|
+
```
|
|
405
|
+
simulstream_inference --speech-processor-config config/speech_processor.yaml \
|
|
406
|
+
--wav-list-file PATH_TO_TXT_FILE_WITH_A_LIST_OF_WAV_FILES.txt \
|
|
407
|
+
--tgt-lang it --src-lang en --metrics-log-file metrics.jsonl
|
|
408
|
+
```
|
|
409
|
+
|
|
410
|
+
Once you have generated the JSONL file containing the inference log (e.g. ``metrics.jsonl``), you
|
|
411
|
+
can score your speech processor by running:
|
|
412
|
+
|
|
413
|
+
```shell
|
|
414
|
+
simulstream_score_latency --scorer stream_laal \
|
|
415
|
+
--eval-config config/speech_processor.yaml \
|
|
416
|
+
--log-file metrics.jsonl \
|
|
417
|
+
--reference REFERENCE_FILE.txt \
|
|
418
|
+
--audio-definition YAML_AUDIO_REFERENCES_DEFINITION.yaml
|
|
419
|
+
|
|
420
|
+
simulstream_score_quality --scorer comet \
|
|
421
|
+
--eval-config config/speech_processor.yaml \
|
|
422
|
+
--log-file metrics.jsonl \
|
|
423
|
+
--references REFERENCES_FILE.txt \
|
|
424
|
+
--transcripts TRANSCRIPTS_FILE.txt
|
|
425
|
+
|
|
426
|
+
simulstream_stats --eval-config config/speech_processor.yaml \
|
|
427
|
+
--log-file metrics.jsonl
|
|
428
|
+
```
|
|
429
|
+
|
|
430
|
+
Each of them will output different metrics. ``simulstream_score_latency`` provides the metric for
|
|
431
|
+
the latency of the system by leveraging a sentence-based TXT reference file and a YAML file, which
|
|
432
|
+
for each line in the reference TXT contains the WAV file to which it belong, its ``offset``
|
|
433
|
+
(i.e. start time) and ``duration``. The exact definition of how the latency is defined depends on
|
|
434
|
+
the selected metric (``--scorer``).
|
|
435
|
+
|
|
436
|
+
Similarly, ``simulstream_score_quality`` evaluated the quality
|
|
437
|
+
of the generated outputs against one (or more) reference (and transcript, only for metrics
|
|
438
|
+
requiring them) file(s).
|
|
439
|
+
|
|
440
|
+
Lastly, ``simulstream_stats`` computes statistics like the computational cost and flickering ratio.
|
|
441
|
+
|
|
442
|
+
|
|
443
|
+
## Contributing
|
|
444
|
+
|
|
445
|
+
Contributions from interested researchers and developers are extremely appreciated.
|
|
446
|
+
|
|
447
|
+
You can create an ***issue*** in case of problems with the code, questions, or feature requests.
|
|
448
|
+
You are also more than welcome to create a ***pull request*** that addresses any ***issue***.
|
|
449
|
+
|
|
450
|
+
## Licence
|
|
451
|
+
|
|
452
|
+
``simulstream`` is licensed under [Apache Version 2.0](LICENSE).
|
|
453
|
+
|
|
454
|
+
## Credits
|
|
455
|
+
If you use this library, please cite:
|
|
456
|
+
|
|
457
|
+
|
|
458
|
+
```
|
|
459
|
+
@article{gaido-et-al-2025-simulstream,
|
|
460
|
+
title={{simulstream: Open-Source Toolkit for Evaluation and Demonstration of Streaming Speech-to-Text Translation Systems}},
|
|
461
|
+
author={Gaido, Marco and Papi, Sara and Cettolo, Mauro and Negri, Matteo and Bentivogli, Luisa},
|
|
462
|
+
journal = "arXiv",
|
|
463
|
+
year={2025}
|
|
464
|
+
}
|
|
465
|
+
```
|
|
@@ -0,0 +1,48 @@
|
|
|
1
|
+
docs/source/conf.py,sha256=aQYxz7x86AkqqgiPnPN8e83RY-nX7D-3WnqYGHRdfmI,1453
|
|
2
|
+
simulstream/__init__.py,sha256=YAGVogIANy5VPzFvbYzs5m1L9PLy5XBu36Oe7dmGR2o,586
|
|
3
|
+
simulstream/config.py,sha256=u9QocK9QgWKgvN5ZrNRv9sIqV0T40wTSC6Ak7VlfoAg,985
|
|
4
|
+
simulstream/inference.py,sha256=YEc4ydAe-lumyUPL-BJD2cfvFmOzP3kEJajjjSN0-yg,6449
|
|
5
|
+
simulstream/client/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
|
6
|
+
simulstream/client/wav_reader_client.py,sha256=1zHNdml9riZxKYM2cyAoUGP1-83XYmMEMioeKBlZH-Y,7861
|
|
7
|
+
simulstream/metrics/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
|
8
|
+
simulstream/metrics/detokenizers.py,sha256=ULkxIOZKwxTx-WLDH1bLJq2ju9BLbR8iaY0Z2tk9BtM,2708
|
|
9
|
+
simulstream/metrics/logger.py,sha256=x7_qN9j9YmyamCct9A1f5WsccyPg6ZCpn-QWXqzhSaY,1093
|
|
10
|
+
simulstream/metrics/readers.py,sha256=PEXm0NJKOeBampRvlAhwrDn7KY1sAuHwqpW6Y893k-0,14469
|
|
11
|
+
simulstream/metrics/score_latency.py,sha256=5A23nkyKbyuJ9Lf_BPQ7g96buEYZNycmY-RqL4JwOPQ,4891
|
|
12
|
+
simulstream/metrics/score_quality.py,sha256=ebiVdgTnCpV1PV0sGsZ0jIs2o3Ekv6ZHpptyxYmeEO4,7185
|
|
13
|
+
simulstream/metrics/stats.py,sha256=HJioWit0lkKBm62uaHb2LaTGs94cwITl301nFUnl9JI,6471
|
|
14
|
+
simulstream/metrics/scorers/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
|
15
|
+
simulstream/metrics/scorers/latency/__init__.py,sha256=nBsATDWJ2m_4o0narJeiI4Ez6maK9BFl-5E3EaqwrxE,3484
|
|
16
|
+
simulstream/metrics/scorers/latency/mwersegmenter.py,sha256=QTHz06SS7ZVVImL9w03YsgrqR-yluQpt2cKfoEx3sz8,5404
|
|
17
|
+
simulstream/metrics/scorers/latency/stream_laal.py,sha256=N_lito76DS72hNhsJKPR5xy9rpmzBwyztb59d-YHsw4,4914
|
|
18
|
+
simulstream/metrics/scorers/quality/__init__.py,sha256=7DlNnAT7Zl6D4zoKMcRJEkkvcC_2lmvYhyR_ftqKiDE,3752
|
|
19
|
+
simulstream/metrics/scorers/quality/comet.py,sha256=eDGMxVtMxKPM1xg36N-91UBb49k9nl1Oim9GYWTUBNI,2131
|
|
20
|
+
simulstream/metrics/scorers/quality/mwersegmenter.py,sha256=RJVJ45t3oCOZNHwhalTkSVnzLwj8ui5juUqy6Aq1TkA,3520
|
|
21
|
+
simulstream/metrics/scorers/quality/sacrebleu.py,sha256=iy7e1ff0BE30xvxDyek8uBt3qJb5DyhLLfHWEP-MrHY,2104
|
|
22
|
+
simulstream/server/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
|
23
|
+
simulstream/server/http_server.py,sha256=6f9uNij6IzxUoQQLIdohhRGDvyTeBwEdGIGYJO5PiaI,3288
|
|
24
|
+
simulstream/server/message_processor.py,sha256=bQFkuMICdKqrxTqYsCpGjhQF-Fdp83XlbB-OMQkkqaA,6648
|
|
25
|
+
simulstream/server/websocket_server.py,sha256=wFcIKUnFbhh6p4BIDpTJevOtOVSf0aYIeAnDX-Ho1NQ,9855
|
|
26
|
+
simulstream/server/speech_processors/__init__.py,sha256=pmDl9h9SmuGGsF2J3SaSjAEaDFre0QWP2F9OsIJQE6s,5377
|
|
27
|
+
simulstream/server/speech_processors/base.py,sha256=y2cFmJiHuJGM7pOJot5xAd-b189V0J164MnWZO9lx7s,4833
|
|
28
|
+
simulstream/server/speech_processors/base_streamatt.py,sha256=bFGsvZyNeHVUpNUcJm4pbeMkCagXCxuv5Ml8Y4VsLjQ,13880
|
|
29
|
+
simulstream/server/speech_processors/canary_sliding_window_retranslation.py,sha256=wJ4KOt48eDobWayABk3GgJS-mhu0gxzlp9T6XjaOqqY,3012
|
|
30
|
+
simulstream/server/speech_processors/hf_sliding_window_retranslation.py,sha256=syqweosf_1xMZNmp3RS_X_EBOrKiHBagxPkhxfMpbUc,3890
|
|
31
|
+
simulstream/server/speech_processors/incremental_output.py,sha256=DVIaHCGTF4v2PzqUHj30Xr-4RtJG6BOqpAMLJY5yG4U,3383
|
|
32
|
+
simulstream/server/speech_processors/seamless_sliding_window_retranslation.py,sha256=G52qTfIm4qiDYUNf1tbdOcJCKE3u-I9KCjpbVSZLorg,3642
|
|
33
|
+
simulstream/server/speech_processors/seamless_streamatt.py,sha256=_Z01FKPJEOSPX2yLHwoBYVebAeagyhSCMbQwnex0h-A,12618
|
|
34
|
+
simulstream/server/speech_processors/simuleval_wrapper.py,sha256=K-gljTmlkJJ9C_RDo34DzNV-dxJdo1xpP3s9mGFzZs0,6655
|
|
35
|
+
simulstream/server/speech_processors/sliding_window_retranslation.py,sha256=h-zbF1Ydy41y34aAZKrmO0NVCh_jGhXYQ1Qdm0Bvxqc,6653
|
|
36
|
+
simulstream/server/speech_processors/vad_wrapper.py,sha256=Ecd0dr_xHxl9kKdswR1-OcEZdPgH9HmEDzz0S6Okrzw,9251
|
|
37
|
+
simulstream-0.1.0.dist-info/licenses/LICENSE,sha256=knu-Hk-hl9xNWU_Gn_6Msa5U8gHhDT8yIHW7_KrmdhE,11333
|
|
38
|
+
uts/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
|
39
|
+
uts/utils.py,sha256=CXC0tahql0HZN4Cb2sQLyYsaW22sDz3n8gCvcVFyx2c,127
|
|
40
|
+
uts/metrics/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
|
41
|
+
uts/metrics/log_reader.py,sha256=6xVFeGDJjl_66juMWl9M5itTZjGv6swowhnpaUwUFcU,2195
|
|
42
|
+
uts/speech_processors/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
|
|
43
|
+
uts/speech_processors/test_simuleval_wrapper.py,sha256=50U9a0DVc-X4ySYsrAooy4wYNM-Uwg0A2PnnMUQNTY8,3517
|
|
44
|
+
simulstream-0.1.0.dist-info/METADATA,sha256=gaFMBeG_4LSuccVtuIaxHKGmxCKyhq_vx5TnduuLXI0,24508
|
|
45
|
+
simulstream-0.1.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
|
|
46
|
+
simulstream-0.1.0.dist-info/entry_points.txt,sha256=5j84rLKVesw47qjaeELpz25XfzuSg_Zz0JLhcO3Y6Gg,479
|
|
47
|
+
simulstream-0.1.0.dist-info/top_level.txt,sha256=DF787K6iu65z7cBQrCRze-l8w8s-ouExTI85-Vxyka8,21
|
|
48
|
+
simulstream-0.1.0.dist-info/RECORD,,
|
|
@@ -0,0 +1,8 @@
|
|
|
1
|
+
[console_scripts]
|
|
2
|
+
simulstream_demo_http_server = simulstream.server.http_server:cli_main
|
|
3
|
+
simulstream_inference = simulstream.inference:cli_main
|
|
4
|
+
simulstream_score_latency = simulstream.metrics.score_latency:cli_main
|
|
5
|
+
simulstream_score_quality = simulstream.metrics.score_quality:cli_main
|
|
6
|
+
simulstream_server = simulstream.server.websocket_server:cli_main
|
|
7
|
+
simulstream_stats = simulstream.metrics.stats:cli_main
|
|
8
|
+
simulstream_wavs_client = simulstream.client.wav_reader_client:cli_main
|