app.py

import openai, os
import gradio as gr
import azure.cognitiveservices.speech as speechsdk
from langchain import OpenAI
from langchain.chains import ConversationChain
from langchain.memory import ConversationSummaryBufferMemory
from langchain.chat_models import ChatOpenAI
from dotenv import load_dotenv
load_dotenv()
# openai.api_base = "https://api.chatanywhere.tech/v1"
# os.environ["OPENAI_API_KEY"]='sk-jfN3iXQqp6DGlhzr6dvxlePzbFa2juVEqQEg8Pi91PrDXNTi'
# #openai.api_key = os.environ["OPENAI_API_KEY"]
# os.environ['AZURE_SPEECH_KEY']='06901dd3964b4681834cfb9bf4fe9efe'
# os.environ['AZURE_SPEECH_REGION']='eastus'


memory = ConversationSummaryBufferMemory(llm=ChatOpenAI(), max_token_limit=2048)
conversation = ConversationChain(
    llm=OpenAI(model_name="gpt-3.5-turbo-0613",max_tokens=2048, temperature=0.5),
    memory=memory,
)

speech_config = speechsdk.SpeechConfig(subscription=os.environ.get('AZURE_SPEECH_KEY'), region=os.environ.get('AZURE_SPEECH_REGION'))
audio_config = speechsdk.audio.AudioOutputConfig(use_default_speaker=True)

# The language of the voice that speaks.
speech_config.speech_synthesis_language='zh-CN'
speech_config.speech_synthesis_voice_name='zh-CN-XiaohanNeural'

speech_synthesizer = speechsdk.SpeechSynthesizer(speech_config=speech_config, audio_config=audio_config)

def play_voice(text):
    speech_synthesizer.speak_text_async(text)

def predict(input, history=[]):
    history.append(input)
    response = conversation.predict(input=input)
    history.append(response)
    play_voice(response)
    responses = [(u,b) for u,b in zip(history[::2], history[1::2])]
    return responses, history

def transcribe(audio):
    os.rename(audio, audio + '.wav')
    audio_file = open(audio + '.wav', "rb")
    transcript = openai.Audio.transcribe("whisper-1", audio_file)
    return transcript['text']

def process_audio(audio, history=[]):
    text = transcribe(audio)
    return predict(text, history)

with gr.Blocks(css="#chatbot{height:800px} .overflow-y-auto{height:800px}") as demo:
    chatbot = gr.Chatbot(elem_id="chatbot")
    state = gr.State([])

    with gr.Row():
        txt = gr.Textbox(show_label=False, placeholder="Enter text and press enter")

    with gr.Row():
        audio = gr.Audio(source="microphone", type="filepath")

    txt.submit(predict, [txt, state], [chatbot, state])
    audio.change(process_audio, [audio, state], [chatbot, state])

demo.launch()