update

main.py

@@ -152,7 +152,7 @@ def main():
                                     ring_max_speech_length_s = gr.Slider(minimum=0.0, maximum=10.0, value=2.0, step=0.05, label="max_speech_length_s")
                                     ring_min_speech_length_s = gr.Slider(minimum=0.0, maximum=2.0, value=0.3, step=0.05, label="min_speech_length_s")
 
-                                ring_button = gr.Button("retrieval", variant="primary")
+                                ring_button = gr.Button("run", variant="primary")
 
                             with gr.Column(scale=1):
                                 ring_image = gr.Image(label="image", height=300, width=720, show_label=False)

toolbox/vad/vad.py

@@ -124,8 +124,8 @@ class Vad(object):
         self.frame_length = int(sample_rate * (frame_length_ms / 1000.0))
         self.frame_step = int(sample_rate * (frame_step_ms / 1000.0))
         self.frame_timestamp_s = 0.0
-        # self.signal_cache = np.zeros(shape=(self.frame_length,), dtype=np.int16)
-        self.signal_cache = None
+        self.signal_cache = np.zeros(shape=(self.frame_length,), dtype=np.int16)
+        # self.signal_cache = None
 
         # segments
         self.num_padding_frames = int(padding_length_ms / frame_step_ms)