dora-vad 0.3.10rc0__tar.gz → 0.3.11rc1__tar.gz

{dora_vad-0.3.10rc0 → dora_vad-0.3.11rc1}/PKG-INFO

@@ -1,12 +1,12 @@
-Metadata-Version: 2.2
+Metadata-Version: 2.4
 Name: dora-vad
-Version: 0.3.10rc0
+Version: 0.3.11rc1
 Summary: Dora Node for Text translating using Argostranslate
 Author-email: Haixuan Xavier Tao <tao.xavier@outlook.com>
 License: MIT
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
-Requires-Dist: dora-rs>=0.3.6
+Requires-Dist: dora-rs>=0.3.9
 Requires-Dist: numpy<2.0.0
 Requires-Dist: silero-vad>=5.1
 

{dora_vad-0.3.10rc0 → dora_vad-0.3.11rc1}/dora_vad/__init__.py

@@ -1,3 +1,5 @@
+"""TODO: Add docstring."""
+
 import os
 
 # Define the path to the README file relative to the package directory

{dora_vad-0.3.10rc0 → dora_vad-0.3.11rc1}/dora_vad/main.py

@@ -1,3 +1,5 @@
+"""TODO: Add docstring."""
+
 import os
 
 import numpy as np
@@ -7,14 +9,15 @@ from dora import Node
 from silero_vad import get_speech_timestamps, load_silero_vad
 
 model = load_silero_vad()
-MIN_SILENCE_DURATION_MS = int(os.getenv("MIN_SILENCE_DURATION_MS", "100"))
+MIN_SILENCE_DURATION_MS = int(os.getenv("MIN_SILENCE_DURATION_MS", "200"))
 MIN_SPEECH_DURATION_MS = int(os.getenv("MIN_SPEECH_DURATION_MS", "300"))
-
-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"))
+THRESHOLD = float(os.getenv("THRESHOLD", "0.4"))
+MAX_AUDIO_DURATION_S = float(os.getenv("MAX_AUDIO_DURATION_S", "75"))
+MIN_AUDIO_SAMPLING_DURATION_MS = int(os.getenv("MIN_AUDIO_SAMPLING_DURATION_MS", "500"))
 
 
 def main():
+    """TODO: Add docstring."""
     node = Node()
     last_audios = []
     while True:
@@ -26,10 +29,11 @@ def main():
             last_audios += [audio]
             last_audios = last_audios[-100:]
             audio = np.concatenate(last_audios)
+            sr = event["metadata"].get("sample_rate", 16000)
             speech_timestamps = get_speech_timestamps(
                 torch.from_numpy(audio),
                 model,
-                threshold=0.2,
+                threshold=THRESHOLD,
                 min_speech_duration_ms=MIN_SPEECH_DURATION_MS,
                 min_silence_duration_ms=MIN_SILENCE_DURATION_MS,
             )
@@ -37,16 +41,20 @@ def main():
             # Check ig there is timestamp
             if (
                 len(speech_timestamps) > 0
-                and len(last_audios) > MIN_AUDIO_SAMPLING_DURAION_S
+                and len(audio) > MIN_AUDIO_SAMPLING_DURATION_MS * sr / 1000
             ):
                 # 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):
+                    node.send_output(
+                        "timestamp_start",
+                        pa.array([speech_timestamps[-1]["start"]]),
+                    )
                     continue
                 audio = audio[0 : speech_timestamps[-1]["end"]]
-                node.send_output("audio", pa.array(audio))
+                node.send_output("audio", pa.array(audio), metadata={"sample_rate": sr})
                 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))
+            elif len(last_audios) > MAX_AUDIO_DURATION_S:
+                node.send_output("audio", pa.array(audio), metadata={"sample_rate": sr})
                 last_audios = []

{dora_vad-0.3.10rc0 → dora_vad-0.3.11rc1}/dora_vad.egg-info/PKG-INFO

@@ -1,12 +1,12 @@
-Metadata-Version: 2.2
+Metadata-Version: 2.4
 Name: dora-vad
-Version: 0.3.10rc0
+Version: 0.3.11rc1
 Summary: Dora Node for Text translating using Argostranslate
 Author-email: Haixuan Xavier Tao <tao.xavier@outlook.com>
 License: MIT
 Requires-Python: >=3.8
 Description-Content-Type: text/markdown
-Requires-Dist: dora-rs>=0.3.6
+Requires-Dist: dora-rs>=0.3.9
 Requires-Dist: numpy<2.0.0
 Requires-Dist: silero-vad>=5.1
 

{dora_vad-0.3.10rc0 → dora_vad-0.3.11rc1}/dora_vad.egg-info/requires.txt

@@ -1,3 +1,3 @@
-dora-rs>=0.3.6
+dora-rs>=0.3.9
 numpy<2.0.0
 silero-vad>=5.1

{dora_vad-0.3.10rc0 → dora_vad-0.3.11rc1}/pyproject.toml

@@ -1,17 +1,28 @@
 [project]
 name = "dora-vad"
-version = "0.3.10-rc0"
+version = "0.3.11-rc1"
 description = "Dora Node for Text translating using Argostranslate"
 authors = [{ name = "Haixuan Xavier Tao", email = "tao.xavier@outlook.com" }]
 license = { text = "MIT" }
 readme = "README.md"
 requires-python = ">=3.8"
 
-dependencies = ["dora-rs >= 0.3.6", "numpy < 2.0.0", "silero-vad >= 5.1"]
-
+dependencies = ["dora-rs >= 0.3.9", "numpy < 2.0.0", "silero-vad >= 5.1"]
 
 [dependency-groups]
 dev = ["pytest >=8.1.1", "ruff >=0.9.1"]
 
 [project.scripts]
 dora-vad = "dora_vad.main:main"
+
+[tool.ruff.lint]
+extend-select = [
+  "D",    # pydocstyle
+  "UP",   # Ruff's UP rule
+  "PERF", # Ruff's PERF rule
+  "RET",  # Ruff's RET rule
+  "RSE",  # Ruff's RSE rule
+  "NPY",  # Ruff's NPY rule
+  "N",    # Ruff's N rule
+  "I",    # Ruff's I rule
+]

{dora_vad-0.3.10rc0 → dora_vad-0.3.11rc1}/tests/test_translate.py

@@ -1,7 +1,10 @@
+"""TODO: Add docstring."""
+
 import pytest
 
 
 def test_import_main():
+    """TODO: Add docstring."""
     from dora_vad.main import main
 
     # Check that everything is working, and catch dora Runtime Exception as we're not running in a dora dataflow.