interfaces/realtime_asr.py

# https://learn.microsoft.com/en-us/azure/cognitive-services/speech-service/get-started-speech-to-text?tabs=linux%2Cterminal&pivots=programming-language-python

import numpy as np
from utils import recognize_from_stream
from azure.cognitiveservices.speech.audio import PushAudioInputStream, AudioStreamFormat

import gradio as gr
import os

import time

stream = PushAudioInputStream(AudioStreamFormat(48000,)) # sample rate is important

msg_queue = []

chat_history = [
]

def rec_cb(evt):
    print("##########################")
    print(evt.result.text)
    if evt.result.text:
        chat_history.append({'role':'user', 'content':evt.result.text})
        
    
speech_recognizer = recognize_from_stream(stream,rec_cb)

def transcribe(speech):
    sample_rate, speech = speech
    print(time.time(), (sample_rate, len(speech)))
    
    stream.write(speech.tobytes())
    
    user_msg = [i['content'] for i in chat_history if i['role']=='user']
    box_msg = [i['content'] for i in chat_history if i['role']=='assistant']
    
    return list(zip(user_msg,box_msg))


demo = gr.Interface(
    title="实时语音识别",
    description="使用Azure的语音识别服务，实时识别麦克风输入的语音。",
    fn=transcribe, 
    inputs=[
        gr.Audio(source="microphone", type="numpy",streaming=True), 
    ], 
    outputs= [
        gr.Chatbot(), 
    ], 
    live=True)

if __name__ == '__main__':
    demo.launch(share=True,show_error=True)