dora-vad 0.3.8__py3-none-any.whl → 0.3.9__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.
- dora_vad/__init__.py +1 -1
- dora_vad/main.py +10 -11
- dora_vad-0.3.9.dist-info/METADATA +57 -0
- dora_vad-0.3.9.dist-info/RECORD +7 -0
- {dora_vad-0.3.8.dist-info → dora_vad-0.3.9.dist-info}/WHEEL +2 -1
- dora_vad-0.3.9.dist-info/entry_points.txt +2 -0
- dora_vad-0.3.9.dist-info/top_level.txt +1 -0
- dora_vad-0.3.8.dist-info/METADATA +0 -24
- dora_vad-0.3.8.dist-info/RECORD +0 -6
- dora_vad-0.3.8.dist-info/entry_points.txt +0 -3
dora_vad/__init__.py
CHANGED
@@ -5,7 +5,7 @@ readme_path = os.path.join(os.path.dirname(os.path.dirname(__file__)), "README.m
|
|
5
5
|
|
6
6
|
# Read the content of the README file
|
7
7
|
try:
|
8
|
-
with open(readme_path,
|
8
|
+
with open(readme_path, encoding="utf-8") as f:
|
9
9
|
__doc__ = f.read()
|
10
10
|
except FileNotFoundError:
|
11
11
|
__doc__ = "README file not found."
|
dora_vad/main.py
CHANGED
@@ -1,13 +1,14 @@
|
|
1
|
-
from dora import Node
|
2
|
-
import pyarrow as pa
|
3
|
-
import numpy as np
|
4
1
|
import os
|
5
|
-
|
2
|
+
|
3
|
+
import numpy as np
|
4
|
+
import pyarrow as pa
|
6
5
|
import torch
|
6
|
+
from dora import Node
|
7
|
+
from silero_vad import get_speech_timestamps, load_silero_vad
|
7
8
|
|
8
9
|
model = load_silero_vad()
|
9
|
-
MIN_SILENCE_DURATION_MS = int(os.getenv("MIN_SILENCE_DURATION_MS", "
|
10
|
-
MIN_SPEECH_DURATION_MS = int(os.getenv("MIN_SPEECH_DURATION_MS", "
|
10
|
+
MIN_SILENCE_DURATION_MS = int(os.getenv("MIN_SILENCE_DURATION_MS", "100"))
|
11
|
+
MIN_SPEECH_DURATION_MS = int(os.getenv("MIN_SPEECH_DURATION_MS", "300"))
|
11
12
|
|
12
13
|
MIN_AUDIO_SAMPLING_DURAION_S = int(os.getenv("MAX_AUDIO_DURATION_S", "20"))
|
13
14
|
MAX_AUDIO_DURAION_S = int(os.getenv("MAX_AUDIO_DURATION_S", "75"))
|
@@ -38,14 +39,12 @@ def main():
|
|
38
39
|
len(speech_timestamps) > 0
|
39
40
|
and len(last_audios) > MIN_AUDIO_SAMPLING_DURAION_S
|
40
41
|
):
|
41
|
-
|
42
42
|
# Check if the audio is not cut at the end. And only return if there is a long time spent
|
43
43
|
if speech_timestamps[-1]["end"] == len(audio):
|
44
44
|
continue
|
45
|
-
|
46
|
-
|
47
|
-
|
48
|
-
last_audios = [audio[speech_timestamps[-1]["end"] :]]
|
45
|
+
audio = audio[0 : speech_timestamps[-1]["end"]]
|
46
|
+
node.send_output("audio", pa.array(audio))
|
47
|
+
last_audios = [audio[speech_timestamps[-1]["end"] :]]
|
49
48
|
|
50
49
|
# If there is no sound for too long return the audio
|
51
50
|
elif len(last_audios) > 75:
|
@@ -0,0 +1,57 @@
|
|
1
|
+
Metadata-Version: 2.2
|
2
|
+
Name: dora-vad
|
3
|
+
Version: 0.3.9
|
4
|
+
Summary: Dora Node for Text translating using Argostranslate
|
5
|
+
Author-email: Haixuan Xavier Tao <tao.xavier@outlook.com>
|
6
|
+
License: MIT
|
7
|
+
Requires-Python: >=3.8
|
8
|
+
Description-Content-Type: text/markdown
|
9
|
+
Requires-Dist: dora-rs>=0.3.6
|
10
|
+
Requires-Dist: numpy<2.0.0
|
11
|
+
Requires-Dist: silero-vad>=5.1
|
12
|
+
|
13
|
+
# Speech Activity Detection(VAD)
|
14
|
+
|
15
|
+
This is using Silero VAD.
|
16
|
+
|
17
|
+
It detects beginning and ending of voice activity within a stream of audio and returns the parts that contains activity.
|
18
|
+
|
19
|
+
There's a maximum amount of voice duration, to avoid having no input for too long.
|
20
|
+
|
21
|
+
## Input/Output Specification
|
22
|
+
|
23
|
+
- inputs:
|
24
|
+
- audio: 8kHz or 16kHz sample rate.
|
25
|
+
- outputs:
|
26
|
+
- audio: Same as input but truncated
|
27
|
+
|
28
|
+
## YAML Specification
|
29
|
+
|
30
|
+
```yaml
|
31
|
+
- id: dora-vad
|
32
|
+
description: Voice activity detection. See; <a href='https://github.com/snakers4/silero-vad'>sidero</a>
|
33
|
+
build: pip install dora-vad
|
34
|
+
path: dora-vad
|
35
|
+
inputs:
|
36
|
+
audio: dora-microphone/audio
|
37
|
+
outputs:
|
38
|
+
- audio
|
39
|
+
```
|
40
|
+
|
41
|
+
## Reference documentation
|
42
|
+
|
43
|
+
- dora-sidero
|
44
|
+
- github: https://github.com/dora-rs/dora/blob/main/node-hub/dora-vad
|
45
|
+
- website: http://dora-rs.ai/docs/nodes/sidero
|
46
|
+
- Sidero
|
47
|
+
- github https://github.com/snakers4/silero-vad
|
48
|
+
|
49
|
+
## Examples
|
50
|
+
|
51
|
+
- Speech to Text
|
52
|
+
- github: https://github.com/dora-rs/dora/blob/main/examples/speech-to-text
|
53
|
+
- website: https://dora-rs.ai/docs/examples/stt
|
54
|
+
|
55
|
+
## License
|
56
|
+
|
57
|
+
The code and model weights are released under the MIT License.
|
@@ -0,0 +1,7 @@
|
|
1
|
+
dora_vad/__init__.py,sha256=HuSK3dnyI9Pb5QAuaKFwQQ3J5SIZnLcKHPJO0norGzc,353
|
2
|
+
dora_vad/main.py,sha256=HfdT87XdtlZfoNRfKbWKl-ylVNFoq9t8acfrzB6n3wE,1913
|
3
|
+
dora_vad-0.3.9.dist-info/METADATA,sha256=B4iuvHVM6YfqEOqkyw-dZ6_kBJU1fjGU--2Ds6Bfq0g,1460
|
4
|
+
dora_vad-0.3.9.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
|
5
|
+
dora_vad-0.3.9.dist-info/entry_points.txt,sha256=X82-C0mpNH9bVjwn24mzfoWX5ssrK6W1Ylk1HPDR0ME,48
|
6
|
+
dora_vad-0.3.9.dist-info/top_level.txt,sha256=1EUgMajnck6tHR4mFagv_uv-t2A6Cabd2wYtEldKm2Q,9
|
7
|
+
dora_vad-0.3.9.dist-info/RECORD,,
|
@@ -0,0 +1 @@
|
|
1
|
+
dora_vad
|
@@ -1,24 +0,0 @@
|
|
1
|
-
Metadata-Version: 2.1
|
2
|
-
Name: dora-vad
|
3
|
-
Version: 0.3.8
|
4
|
-
Summary: Dora Node for Text translating using Argostranslate
|
5
|
-
Author: Haixuan Xavier Tao
|
6
|
-
Author-email: tao.xavier@outlook.com
|
7
|
-
Requires-Python: >=3.7,<4.0
|
8
|
-
Classifier: Programming Language :: Python :: 3
|
9
|
-
Classifier: Programming Language :: Python :: 3.7
|
10
|
-
Classifier: Programming Language :: Python :: 3.8
|
11
|
-
Classifier: Programming Language :: Python :: 3.9
|
12
|
-
Classifier: Programming Language :: Python :: 3.10
|
13
|
-
Classifier: Programming Language :: Python :: 3.11
|
14
|
-
Classifier: Programming Language :: Python :: 3.12
|
15
|
-
Classifier: Programming Language :: Python :: 3.13
|
16
|
-
Requires-Dist: dora-rs (>=0.3.6,<0.4.0)
|
17
|
-
Requires-Dist: numpy (<2.0.0)
|
18
|
-
Requires-Dist: silero-vad (>=5.1,<6.0)
|
19
|
-
Description-Content-Type: text/markdown
|
20
|
-
|
21
|
-
# Speech Activity Detection(VAD)
|
22
|
-
|
23
|
-
This is using Silero VAD.
|
24
|
-
|
dora_vad-0.3.8.dist-info/RECORD
DELETED
@@ -1,6 +0,0 @@
|
|
1
|
-
dora_vad/__init__.py,sha256=Gy4qL4vCeTyA5HR1Yp3ioL4-ClJyW8oi_38CzMuMsBM,358
|
2
|
-
dora_vad/main.py,sha256=vCTRvRcn51P9gBznRaFNtdz1rSycQDag66bm9Eficzw,1948
|
3
|
-
dora_vad-0.3.8.dist-info/METADATA,sha256=FZwAgVoMSzR-2bxlIOK9xVrY_8MRHbzoO-cr12-FTOA,818
|
4
|
-
dora_vad-0.3.8.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
|
5
|
-
dora_vad-0.3.8.dist-info/entry_points.txt,sha256=wr9gBgtZsPnfZYt4HJD6Pf7if_ieGubTZxuEbWbJmQI,47
|
6
|
-
dora_vad-0.3.8.dist-info/RECORD,,
|