can detect sentece ends

app.py

@@ -2,6 +2,7 @@ import gradio as gr
 import numpy as np
 from transcriber import AudioProcessor
 
+
 # Create processor instance with more conservative settings
 processor = AudioProcessor(model_size="tiny.en", device="cpu")
 
@@ -22,6 +23,8 @@ def process_mic_audio(audio):
     # Get current transcription
     transcription = processor.get_transcription()
     print(transcription)
+    transcription = str(transcription)
+
 
     # Return status update and transcription
     buffer_seconds = buffer_size / processor.sample_rate

transcriber.py

@@ -3,6 +3,8 @@ import threading
 import time
 from faster_whisper import WhisperModel
 import scipy.signal as signal
+from typing import List
+from punctuators.models import SBDModelONNX
 
 class AudioProcessor:
     def __init__(self, model_size="tiny.en", device="cpu", compute_type="int8"):
@@ -26,6 +28,8 @@ class AudioProcessor:
         self.audio_model = WhisperModel(model_size, device=device, compute_type=compute_type)
         print(f"Initialized {model_size} model on {device}")
 
+        self.sentence_end_detect = SBDModelONNX.from_pretrained("sbd_multi_lang")
+
     def _trim_buffer_intelligently(self):
         """
         Trim the buffer while preserving transcription continuity
@@ -273,7 +277,8 @@ class AudioProcessor:
     def get_transcription(self):
         """Get the current transcription text"""
         with self.lock:
-            return self.full_transcription
+            results: List[List[str]] = self.sentence_end_detect.infer([self.full_transcription])
+            return results[0]
 
     def get_playback_audio(self):
         """Get properly formatted audio for Gradio playback"""

working.py

@@ -1,28 +0,0 @@
-transcriber = pipeline("automatic-speech-recognition", model="openai/whisper-base.en")
-
-def transcribe(stream, new_chunk):
-    sr, y = new_chunk
-    
-    # Convert to mono if stereo
-    if y.ndim > 1:
-        y = y.mean(axis=1)
-        
-    y = y.astype(np.float32)
-    y /= np.max(np.abs(y))
-
-    if stream is not None:
-        stream = np.concatenate([stream, y])
-    else:
-        stream = y
-    
-    # Return the stream as state and a string representation of the array for display
-    return stream, str(stream)
-
-demo = gr.Interface(
-    transcribe,
-    ["state", gr.Audio(sources=["microphone"], streaming=True)],
-    ["state", "text"],
-    live=True,
-)
-
-demo.launch()