edge-gemma-speak 0.1.0__py3-none-any.whl → 0.1.1__py3-none-any.whl

edge_gemma_speak/__init__.py

@@ -12,7 +12,7 @@ from .voice_assistant import (
     main
 )
 
-__version__ = "0.1.0"
+__version__ = "0.1.1"
 __author__ = "MimicLab, Sogang University"
 
 __all__ = [

edge_gemma_speak/cli.py

@@ -107,6 +107,10 @@ def main():
                        help="STT temperature for sampling (default: 0.0)")
     parser.add_argument("--stt-vad-threshold", type=float, default=0.5,
                        help="STT VAD threshold (default: 0.5)")
+    parser.add_argument("--stt-vad-min-speech-duration", type=int, default=250,
+                       help="Minimum speech duration in ms (default: 250)")
+    parser.add_argument("--stt-vad-min-silence-duration", type=int, default=1000,
+                       help="Minimum silence duration in ms before cutting off (default: 1000)")
     
     # LLM 파라미터
     parser.add_argument("--llm-max-tokens", type=int, default=512,
@@ -268,6 +272,8 @@ def main():
         stt_beam_size=args.stt_beam_size,
         stt_temperature=args.stt_temperature,
         stt_vad_threshold=args.stt_vad_threshold,
+        stt_vad_min_speech_duration_ms=args.stt_vad_min_speech_duration,
+        stt_vad_min_silence_duration_ms=args.stt_vad_min_silence_duration,
         # TTS parameters
         tts_voice=tts_voice,
         # LLM parameters

edge_gemma_speak/voice_assistant.py

@@ -72,7 +72,7 @@ class ModelConfig:
     stt_temperature: float = 0.0
     stt_vad_threshold: float = 0.5
     stt_vad_min_speech_duration_ms: int = 250
-    stt_vad_min_silence_duration_ms: int = 2000
+    stt_vad_min_silence_duration_ms: int = 1000  # Reduced from 2000ms for faster response
     
     # TTS detailed settings
     tts_voice: str = "ko-KR-HyunsuMultilingualNeural"

{edge_gemma_speak-0.1.0.dist-info → edge_gemma_speak-0.1.1.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: edge_gemma_speak
-Version: 0.1.0
+Version: 0.1.1
 Summary: Edge-based voice assistant using Gemma LLM with STT and TTS capabilities
 Home-page: https://github.com/yourusername/edge_gemma_speak
 Author: MimicLab, Sogang University
@@ -34,6 +34,9 @@ Dynamic: license-file
 Dynamic: requires-python
 
 # 🎙️ Edge Gemma Speak
+[![PyPI Status](https://badge.fury.io/py/edge-gemma-speak.svg)](https://badge.fury.io/py/edge-gemma-speak)
+[![license](https://img.shields.io/badge/License-MIT-blue.svg)](https://github.com/MIMICLab/EdgeGemmaSpeak/blob/main/LICENSE)
+[![Downloads](https://pepy.tech/badge/edge-gemma-speak)](https://pepy.tech/project/edge-gemma-speak)
 
 Edge-based voice assistant using Gemma LLM with Speech-to-Text and Text-to-Speech capabilities
 
@@ -122,6 +125,12 @@ edge-gemma-speak --stt-beam-size 10
 # Adjust VAD sensitivity (default: 0.5)
 edge-gemma-speak --stt-vad-threshold 0.3
 
+# Adjust minimum speech duration in ms (default: 250)
+edge-gemma-speak --stt-vad-min-speech-duration 200
+
+# Adjust minimum silence duration in ms (default: 1000)
+edge-gemma-speak --stt-vad-min-silence-duration 800
+
 # Change Whisper model size (tiny, base, small, medium, large)
 edge-gemma-speak --stt-model small
 ```
@@ -296,6 +305,7 @@ For large LLM models:
 - Check microphone permissions in system settings
 - Close other audio applications
 - Adjust VAD threshold: `--stt-vad-threshold 0.3`
+- Reduce silence duration for faster response: `--stt-vad-min-silence-duration 500`
 
 ### Model File Not Found
 

edge_gemma_speak-0.1.1.dist-info/RECORD

@@ -0,0 +1,9 @@
+edge_gemma_speak/__init__.py,sha256=D2nUTwXZlFuNG0XVpfQSs3pdwmS54FjqrhjWBQzjSzE,445
+edge_gemma_speak/cli.py,sha256=sn7nLQtJlNE7lL-qCGRBTRj0hOj6Ec0r6mzxUlE2aFI,13001
+edge_gemma_speak/voice_assistant.py,sha256=1m_3k-d1LSY1qoSxTKD9IL0qbxkpfkLcW0M3asBAN5U,26535
+edge_gemma_speak-0.1.1.dist-info/licenses/LICENSE,sha256=PYqztu9Bfz2u_QNatdRVQ-K2_ywpSgwFiGiKrqQdayQ,1083
+edge_gemma_speak-0.1.1.dist-info/METADATA,sha256=sJKbbg8yuyJ1_kGfl0DhXKlvXn7vCAhYQg4WfYMWDDs,10230
+edge_gemma_speak-0.1.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+edge_gemma_speak-0.1.1.dist-info/entry_points.txt,sha256=nJgB49bs1CJFwrZkK_x4MBBNo8l8Ihf3EBGpBvr94pU,63
+edge_gemma_speak-0.1.1.dist-info/top_level.txt,sha256=JUyz2j-g_v8aA-lWzxlBHaMpzZTvMR9MJH-h1ye3wa8,17
+edge_gemma_speak-0.1.1.dist-info/RECORD,,

edge_gemma_speak-0.1.0.dist-info/RECORD

@@ -1,9 +0,0 @@
-edge_gemma_speak/__init__.py,sha256=HPS-2No6K2IyCSnK0wLUci0mN-ZSlmgSjYw9nGLPUfA,445
-edge_gemma_speak/cli.py,sha256=99N4k-U9ORpcXtKOPYd3OI4SCBmZK-ayM39hvXM_2zY,12518
-edge_gemma_speak/voice_assistant.py,sha256=xR8Mn75EOZuFeOn9XHruaIIAo1qu5jOj9-NYEuX8VSQ,26492
-edge_gemma_speak-0.1.0.dist-info/licenses/LICENSE,sha256=PYqztu9Bfz2u_QNatdRVQ-K2_ywpSgwFiGiKrqQdayQ,1083
-edge_gemma_speak-0.1.0.dist-info/METADATA,sha256=Z_X-P8mE7VbR1j4yU12tsXnZ2-RFY76TCoTlMXn96-A,9597
-edge_gemma_speak-0.1.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-edge_gemma_speak-0.1.0.dist-info/entry_points.txt,sha256=nJgB49bs1CJFwrZkK_x4MBBNo8l8Ihf3EBGpBvr94pU,63
-edge_gemma_speak-0.1.0.dist-info/top_level.txt,sha256=JUyz2j-g_v8aA-lWzxlBHaMpzZTvMR9MJH-h1ye3wa8,17
-edge_gemma_speak-0.1.0.dist-info/RECORD,,