dora-vad 0.3.7__py3-none-any.whl

dora_vad/__init__.py

@@ -0,0 +1,11 @@
+import os
+
+# Define the path to the README file relative to the package directory
+readme_path = os.path.join(os.path.dirname(os.path.dirname(__file__)), "README.md")
+
+# Read the content of the README file
+try:
+    with open(readme_path, "r", encoding="utf-8") as f:
+        __doc__ = f.read()
+except FileNotFoundError:
+    __doc__ = "README file not found."

dora_vad/main.py

@@ -0,0 +1,53 @@
+from dora import Node
+import pyarrow as pa
+import numpy as np
+import os
+from silero_vad import load_silero_vad, get_speech_timestamps
+import torch
+
+model = load_silero_vad()
+MIN_SILENCE_DURATION_MS = int(os.getenv("MIN_SILENCE_DURATION_MS", "200"))
+MIN_SPEECH_DURATION_MS = int(os.getenv("MIN_SPEECH_DURATION_MS", "1000"))
+
+MIN_AUDIO_SAMPLING_DURAION_S = int(os.getenv("MAX_AUDIO_DURATION_S", "20"))
+MAX_AUDIO_DURAION_S = int(os.getenv("MAX_AUDIO_DURATION_S", "75"))
+
+
+def main():
+    node = Node()
+    last_audios = []
+    while True:
+        event = node.next()
+        if event is None:
+            break
+        if event["type"] == "INPUT" and event["id"] == "audio":
+            audio = event["value"].to_numpy()
+            last_audios += [audio]
+            last_audios = last_audios[-100:]
+            audio = np.concatenate(last_audios)
+            speech_timestamps = get_speech_timestamps(
+                torch.from_numpy(audio),
+                model,
+                threshold=0.2,
+                min_speech_duration_ms=MIN_SPEECH_DURATION_MS,
+                min_silence_duration_ms=MIN_SILENCE_DURATION_MS,
+            )
+
+            # Check ig there is timestamp
+            if (
+                len(speech_timestamps) > 0
+                and len(last_audios) > MIN_AUDIO_SAMPLING_DURAION_S
+            ):
+
+                # Check if the audio is not cut at the end. And only return if there is a long time spent
+                if speech_timestamps[-1]["end"] == len(audio):
+                    continue
+                else:
+                    audio = audio[0 : speech_timestamps[-1]["end"]]
+                    node.send_output("audio", pa.array(audio))
+                    last_audios = [audio[speech_timestamps[-1]["end"] :]]
+
+            # If there is no sound for too long return the audio
+            elif len(last_audios) > 75:
+                node.send_output("audio", pa.array(audio))
+                last_audios = []

dora_vad-0.3.7.dist-info/METADATA

@@ -0,0 +1,24 @@
+Metadata-Version: 2.1
+Name: dora-vad
+Version: 0.3.7
+Summary: Dora Node for Text translating using Argostranslate
+Author: Haixuan Xavier Tao
+Author-email: tao.xavier@outlook.com
+Requires-Python: >=3.7,<4.0
+Classifier: Programming Language :: Python :: 3
+Classifier: Programming Language :: Python :: 3.7
+Classifier: Programming Language :: Python :: 3.8
+Classifier: Programming Language :: Python :: 3.9
+Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
+Classifier: Programming Language :: Python :: 3.13
+Requires-Dist: dora-rs (>=0.3.6,<0.4.0)
+Requires-Dist: numpy (<2.0.0)
+Requires-Dist: silero-vad (>=5.1,<6.0)
+Description-Content-Type: text/markdown
+
+# Speech Activity Detection(VAD)
+
+This is using Silero VAD.
+

dora_vad-0.3.7.dist-info/RECORD

@@ -0,0 +1,6 @@
+dora_vad/__init__.py,sha256=Gy4qL4vCeTyA5HR1Yp3ioL4-ClJyW8oi_38CzMuMsBM,358
+dora_vad/main.py,sha256=vCTRvRcn51P9gBznRaFNtdz1rSycQDag66bm9Eficzw,1948
+dora_vad-0.3.7.dist-info/METADATA,sha256=c9fXyxbuHxeH96JrUa0xdKDnlYcpm-A8DV4P4qbPTik,818
+dora_vad-0.3.7.dist-info/WHEEL,sha256=Nq82e9rUAnEjt98J6MlVmMCZb-t9cYE2Ir1kpBmnWfs,88
+dora_vad-0.3.7.dist-info/entry_points.txt,sha256=wr9gBgtZsPnfZYt4HJD6Pf7if_ieGubTZxuEbWbJmQI,47
+dora_vad-0.3.7.dist-info/RECORD,,

dora_vad-0.3.7.dist-info/WHEEL

@@ -0,0 +1,4 @@
+Wheel-Version: 1.0
+Generator: poetry-core 1.9.1
+Root-Is-Purelib: true
+Tag: py3-none-any

dora_vad-0.3.7.dist-info/entry_points.txt

@@ -0,0 +1,3 @@
+[console_scripts]
+dora-vad=dora_vad.main:main
+