adding post processing

amondin/main.py

@@ -47,7 +47,7 @@ def transcribe(
     }
 
     print("Segmenting speakers...")
-    speaker_segments = segment_speakers(
+    segments = segment_speakers(
         audio,
         hf_token=hf_token,
         num_speakers=num_speakers,
@@ -58,6 +58,13 @@ def transcribe(
     )
 
     print("Transcribing audio...")
+    transcript = speech2text(
+        [segment["audio"] for segment in segments],
+        model_name=s2t_model,
+        language=language,
+        device=device
+    )
+    """
     transcript = []
     for i, speaker_section in enumerate(speaker_segments):
         print(f"Transcribing part {i+1} of {len(speaker_segments)}")
@@ -71,7 +78,7 @@ def transcribe(
         transcript.append(
             [speaker_section["speaker"], speaker_section["time_stamp"], text]
         )
-
+    """
     # Store transcript in pandas Data Frame
     transcript = pd.DataFrame(data=transcript, columns=["speaker", "time_stamp", "text"])
 

amondin/segment_speakers.py

@@ -52,7 +52,6 @@ def segment_speakers(
     # store all passages in a list of dicts
     speaker_segments = []
     for segment in segments:
-        print(segment.duration)
         if segment.duration > 0.1:
             # get audio passages as numpy array
             waveform, sample_rate = Audio().crop(audio, segment, mode="pad")

amondin/speech2text.py

@@ -44,7 +44,7 @@ def speech2text(
         device=device,
     )
 
-    result = pipe(
+    results = pipe(
         audio,
         generate_kwargs={
             "task": "transcribe",
@@ -53,4 +53,5 @@ def speech2text(
     )
 
     # return sting in list
-    return result["text"]
+    return [result["text"] for result in results]
+    return results["text"]