add the azure ttl

app.py

@@ -39,7 +39,8 @@ from langchain.text_splitter import CharacterTextSplitter
 from langchain.vectorstores.faiss import FAISS
 from langchain.docstore.document import Document
 from langchain.chains.question_answering import load_qa_chain
-
+import azure.cognitiveservices.speech as speechsdk
+ 
 
 news_api_key = os.environ["NEWS_API_KEY"]
 
@@ -434,9 +435,9 @@ class ChatWrapper:
                     else:
                         temp_file = LOOPING_TALKING_HEAD
                         html_video = create_html_video(temp_file, TALKING_HEAD_WIDTH)
-                        html_audio, temp_aud_file = do_html_audio_speak(output, translate_to)
+                        html_audio, temp_aud_file = do_html_audio_speak_azure(output, translate_to)
                 else:
-                    html_audio, temp_aud_file = do_html_audio_speak(output, translate_to)
+                    html_audio, temp_aud_file = do_html_audio_speak_azure(output, translate_to)
             else:
                 if talking_head:
                     temp_file = LOOPING_TALKING_HEAD
@@ -456,6 +457,50 @@ class ChatWrapper:
 
 chat = ChatWrapper()
 
+def do_html_audio_speak_azure(words_to_speak, axure_language):
+    speech_key=os.environ["SPEECH_KEY"],
+    service_region=os.environ["SERVICE_REGION"],
+
+
+    speech_config = speechsdk.SpeechConfig(subscription=speech_key, region=service_region)
+    # Note: the voice setting will not overwrite the voice element in input SSML.
+    speech_config.speech_synthesis_voice_name = "zh-CN-XiaoxiaoNeural"
+
+    # 设置输出的音频文件路径和文件名
+    audio_config = speechsdk.audio.AudioOutputConfig(filename="audios/tempfile.mp3")
+
+    text = words_to_speak
+
+    # use the default speaker as audio output.
+    speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=audio_config)
+
+    result = speech_synthesizer.speak_text_async(text).get()
+    # Check result
+    if result.reason == speechsdk.ResultReason.SynthesizingAudioCompleted:
+        print("Speech synthesized for text [{}]".format(text))
+        try:
+            with open('audios/tempfile.mp3', 'wb') as f:
+                f.write(stream.read())
+            temp_aud_file = gr.File("audios/tempfile.mp3")
+            temp_aud_file_url = "/file=" + temp_aud_file.value['name']
+            html_audio = f'<audio autoplay><source src={temp_aud_file_url} type="audio/mp3"></audio>'
+        except IOError as error:
+            # Could not write to file, exit gracefully
+            print(error)
+            return None, None
+    elif result.reason == speechsdk.ResultReason.Canceled:
+        cancellation_details = result.cancellation_details
+        print("Speech synthesis canceled: {}".format(cancellation_details.reason))
+        if cancellation_details.reason == speechsdk.CancellationReason.Error:
+            print("Error details: {}".format(cancellation_details.error_details))
+        # The response didn't contain audio data, exit gracefully
+        print("Could not stream audio")
+        return None, None
+
+    html_audio = '<pre>no audio</pre>'
+
+    return html_audio, "audios/tempfile.mp3"
+
 
 def do_html_audio_speak(words_to_speak, polly_language):
     polly_client = boto3.Session(

requirements.txt

@@ -8,3 +8,4 @@ requests==2.28.2
 git+https://github.com/openai/whisper.git
 boto3==1.26.93
 faiss-cpu
+azure-cognitiveservices-speech==1.27.0