change polly to real time

app.py

@@ -125,61 +125,63 @@ def download_file(bucket_name, object_key, file_path):
     except Exception as e:
         print(f"Error downloading file: {str(e)}")
 
+def polly_text_to_audio(audio_file_name, text, audio_format):
 
-def play_s3_voice(text):
-    response = polly.start_speech_synthesis_task(
-        OutputS3BucketName='lingo-audio-materials', #this bucket is in us-east-1
-        OutputS3KeyPrefix='answers/',
-        OutputFormat='mp3',
+    polly_response = polly.synthesize_speech(
         Text=text,
+        OutputFormat=audio_format,  
+        SampleRate='16000',
         VoiceId='Zhiyu',
         LanguageCode='cmn-CN',
-        Engine='neural'
-    )
-    
-    # Print the task ID and status
-    task_id = response['SynthesisTask']['TaskId']
-    print('Task ID:', task_id)
-    
-    while True:
-        task = polly.get_speech_synthesis_task(TaskId=task_id)
-        task_status = task['SynthesisTask']['TaskStatus']
+        Engine='neural',
+        LexiconNames=['tigoCN']
+    )   
     
-        if task_status == 'completed':
-            break
-        elif task_status  == 'failed':
-            # Task failed
-            print('Task failed:', task['SynthesisTask']['TaskStatusReason'])
-            break
-        else:
-            print("Polly synthesis task is still in progress...")
-            time.sleep(1)
+    # Access the audio stream from the response
+    if "AudioStream" in polly_response:
+        # Note: Closing the stream is important because the service throttles on the
+        # number of parallel connections. Here we are using contextlib.closing to
+        # ensure the close method of the stream object will be called automatically
+        # at the end of the with statement's scope.
+            with closing(polly_response["AudioStream"]) as stream:
+               try:
+                # Open a file for writing the output as a binary stream
+                    with open(audio_file_name, "ab") as file:
+                       file.write(stream.read())
+               except IOError as error:
+                  # Could not write to file, exit gracefully
+                  print(error)
+                  sys.exit(-1)
+
+    else:
+        # The response didn't contain audio data, exit gracefully
+        print("Could not stream audio")
+        sys.exit(-1)
+
+
+def play_s3_voice(text):
+
+    output_file = "/tmp/response.mp3"
+    polly_text_to_audio(output_file, text, "mp3")
 
-    output_uri = response['SynthesisTask']['OutputUri']
-    print("polly output_uri:"+output_uri)
-    output_uri = output_uri.replace("https://", "")
+    # Upload the file to an S3 bucket
+    audio_output_bucket_name = "lingo-audio-materials"
+    audio_output_s3_key = "answers/response.mp3"
 
-    # Split the URI into bucket name and key
-    results = output_uri.split("/")
-    bucket_name = results[1]
-    key_name = results[2]+'/'+results[3]
-    print("bucket name:"+bucket_name)
-    print("key name:"+key_name)
+    s3.upload_file(file_name, audio_output_bucket_name, audio_output_s3_key)
+    
+    # Construct the S3 bucket URI
+    s3_uri = f"s3://{audio_output_bucket_name}/{audio_output_s3_key}"
 
-    mp3_pre_signed_url = s3.generate_presigned_url('get_object',Params={'Bucket': bucket_name,'Key': key_name},ExpiresIn=3600)
+    print("audio output bucket name:"+audio_output_bucket_name)
+    print("audio output key name:"+audio_output_s3_key)
+
+    mp3_pre_signed_url = s3.generate_presigned_url('get_object',Params={'Bucket': audio_output_bucket_name,'Key': audio_output_s3_key},ExpiresIn=3600)
     print("mp3_pre_signed_url:"+mp3_pre_signed_url)
 
-    output_file = "/tmp/from-s3.mp3"
     current_dir = os.getcwd()
-    #file_absolute_path = current_dir+'/'+output_file
     print("current dir:"+current_dir)
     print("output_file_location: "+output_file)
-    download_file(bucket_name, key_name, output_file)
-    #encoded_path = file_absolute_path.encode("utf-8")
-
-    #tmp_aud_file_url = output_file
-    #htm_audio = f'<audio><source src={tmp_aud_file_url} type="audio/mp3" autoplay></audio>'
-    #audio_htm = gr.HTML(htm_audio)
     return output_file, mp3_pre_signed_url
 
 def generate_talk_with_audio(input, avatar_url, api_key = did_api_key):