Add get_buffer method to retrieve unvalidated buffer in string format

src/whisper_streaming/online_asr.py

@@ -85,6 +85,7 @@ class HypothesisBuffer:
             self.committed_in_buffer.pop(0)
 
 
+
 class OnlineASRProcessor:
     """
     Processes incoming audio in a streaming fashion, calling the ASR system
@@ -163,6 +164,13 @@ class OnlineASRProcessor:
         context_text = self.asr.sep.join(token.text for token in non_prompt_tokens)
         return self.asr.sep.join(prompt_list[::-1]), context_text
 
+    def get_buffer(self):
+        """
+        Get the unvalidated buffer in string format.
+        """
+        return self.concatenate_tokens(self.transcript_buffer.buffer).text
+        
+
     def process_iter(self) -> Transcript:
         """
         Processes the current audio buffer.
@@ -414,3 +422,9 @@ class VACOnlineASRProcessor:
         self.current_online_chunk_buffer_size = 0
         self.is_currently_final = False
         return result
+    
+    def get_buffer(self):
+        """
+        Get the unvalidated buffer in string format.
+        """
+        return self.online.concatenate_tokens(self.online.transcript_buffer.buffer).text

whisper_fastapi_online_server.py

@@ -158,12 +158,8 @@ async def websocket_endpoint(websocket: WebSocket):
                         })
                     
                     full_transcription += transcription.text
-                    if args.vac:
+                    buffer = online.get_buffer()
-                        transcript = online.online.concatenate_tokens(online.online.transcript_buffer.buffer)
-                    else:
-                        transcript = online.concatenate_tokens(online.transcript_buffer.buffer)
                   
-                    buffer = transcript.text  
                     if buffer in full_transcription: # With VAC, the buffer is not updated until the next chunk is processed
                         buffer = ""