提交

kongdeqiang

8 天以前 28ccfbfc51068a663a80764e14074df5edf2b5ba

 runtime/python/websocket/funasr_client_api.py

@@ -1,9 +1,10 @@
'''

"""

  Copyright FunASR (https://github.com/alibaba-damo-academy/FunASR). All Rights

  Reserved. MIT License  (https://opensource.org/licenses/MIT)

  

  2022-2023 by zhaomingwork@qq.com  

'''

"""



# pip install websocket-client

import ssl

from websocket import ABNF

@@ -14,138 +15,146 @@
import json

import time

import numpy as np





# class for recognizer in websocket

class Funasr_websocket_recognizer():

    '''

class Funasr_websocket_recognizer:

    """

    python asr recognizer lib



    '''

    def __init__(self, host="127.0.0.1",

                 port="30035",

                 is_ssl=True,

                 chunk_size="0, 10, 5",

                 chunk_interval=10,

                 mode="offline",

                 wav_name="default"):

      '''

          host: server host ip

          port: server port

          is_ssl: True for wss protocal, False for ws

      '''

      try:

        if is_ssl == True:

            ssl_context = ssl.SSLContext()

            ssl_context.check_hostname = False

            ssl_context.verify_mode = ssl.CERT_NONE

            uri = "wss://{}:{}".format(host, port)

            ssl_opt={"cert_reqs": ssl.CERT_NONE}

        else:

            uri = "ws://{}:{}".format(host, port)

            ssl_context = None

            ssl_opt=None

        self.host = host

        self.port = port

 
        self.msg_queue = Queue() # used for recognized result text

    """



        print("connect to url",uri)

        self.websocket=create_connection(uri, ssl=ssl_context, sslopt=ssl_opt)

 
        self.thread_msg = threading.Thread(target=Funasr_websocket_recognizer.thread_rec_msg, args=(self,))

        self.thread_msg.start()

        chunk_size = [int(x) for x in  chunk_size.split(",")]

        stride = int(60 *  chunk_size[1] / chunk_interval / 1000 * 16000 * 2)

        chunk_num = (len(audio_bytes) - 1) // stride + 1

       
        message = json.dumps({"mode": mode,

                              "chunk_size": chunk_size,

                              "encoder_chunk_look_back": 4,

                              "decoder_chunk_look_back": 1,

                              "chunk_interval": chunk_interval,

                              "wav_name": wav_name,

                              "is_speaking": True})

 
        self.websocket.send(message)

 
        print("send json",message)

      
      except Exception as e:

    def __init__(

        self,

        host="127.0.0.1",

        port="30035",

        is_ssl=True,

        chunk_size="0, 10, 5",

        chunk_interval=10,

        mode="offline",

        wav_name="default",

    ):

        """

        host: server host ip

        port: server port

        is_ssl: True for wss protocal, False for ws

        """

        try:

            if is_ssl == True:

                ssl_context = ssl.SSLContext()

                ssl_context.check_hostname = False

                ssl_context.verify_mode = ssl.CERT_NONE

                uri = "wss://{}:{}".format(host, port)

                ssl_opt = {"cert_reqs": ssl.CERT_NONE}

            else:

                uri = "ws://{}:{}".format(host, port)

                ssl_context = None

                ssl_opt = None

            self.host = host

            self.port = port



            self.msg_queue = Queue()  # used for recognized result text



            print("connect to url", uri)

            self.websocket = create_connection(uri, ssl=ssl_context, sslopt=ssl_opt)



            self.thread_msg = threading.Thread(

                target=Funasr_websocket_recognizer.thread_rec_msg, args=(self,)

            )

            self.thread_msg.start()

            chunk_size = [int(x) for x in chunk_size.split(",")]

            stride = int(60 * chunk_size[1] / chunk_interval / 1000 * 16000 * 2)

            chunk_num = (len(audio_bytes) - 1) // stride + 1



            message = json.dumps(

                {

                    "mode": mode,

                    "chunk_size": chunk_size,

                    "encoder_chunk_look_back": 4,

                    "decoder_chunk_look_back": 1,

                    "chunk_interval": chunk_interval,

                    "wav_name": wav_name,

                    "is_speaking": True,

                }

            )



            self.websocket.send(message)



            print("send json", message)



        except Exception as e:

            print("Exception:", e)

            traceback.print_exc()

    


    # threads for rev msg

    def thread_rec_msg(self):

        try:

         while(True):

           msg=self.websocket.recv()

           if msg is None or len(msg) == 0:

             continue

           msg = json.loads(msg)

           
           self.msg_queue.put(msg)

            while True:

                msg = self.websocket.recv()

                if msg is None or len(msg) == 0:

                    continue

                msg = json.loads(msg)



                self.msg_queue.put(msg)

        except Exception as e:

            print("client closed")

 


    # feed data to asr engine, wait_time means waiting for result until time out

    def feed_chunk(self, chunk, wait_time=0.01):

        try:

            self.websocket.send(chunk,  ABNF.OPCODE_BINARY)

            self.websocket.send(chunk, ABNF.OPCODE_BINARY)

            # loop to check if there is a message, timeout in 0.01s

            while(True):

               msg = self.msg_queue.get(timeout=wait_time)

               if self.msg_queue.empty():

                  break

                  
            while True:

                msg = self.msg_queue.get(timeout=wait_time)

                if self.msg_queue.empty():

                    break



            return msg

        except:

            return ""

        
    def close(self,timeout=1):



    def close(self, timeout=1):

        message = json.dumps({"is_speaking": False})

        self.websocket.send(message)

        # sleep for timeout seconds to wait for result

        time.sleep(timeout)

        msg=""

        while(not self.msg_queue.empty()):

        msg = ""

        while not self.msg_queue.empty():

            msg = self.msg_queue.get()

        


        self.websocket.close()

        # only resturn the last msg

        return msg

        
if __name__ == '__main__':

    
    print('example for Funasr_websocket_recognizer') 




if __name__ == "__main__":



    print("example for Funasr_websocket_recognizer")

    import wave



    wav_path = "/Users/zhifu/Downloads/modelscope_models/speech_seaco_paraformer_large_asr_nat-zh-cn-16k-common-vocab8404-pytorch/example/asr_example.wav"

    with wave.open(wav_path, "rb") as wav_file:

                params = wav_file.getparams()

                frames = wav_file.readframes(wav_file.getnframes())

                audio_bytes = bytes(frames)

    
 
        params = wav_file.getparams()

        frames = wav_file.readframes(wav_file.getnframes())

        audio_bytes = bytes(frames)



    stride = int(60 * 10 / 10 / 1000 * 16000 * 2)

    chunk_num = (len(audio_bytes) - 1) // stride + 1

    # create an recognizer 
    rcg = Funasr_websocket_recognizer(host="127.0.0.1",

                                      port="10095",

                                      is_ssl=True,

                                      mode="2pass",

                                      chunk_size="0,10,5")

    # create an recognizer

    rcg = Funasr_websocket_recognizer(

        host="127.0.0.1", port="10095", is_ssl=True, mode="2pass", chunk_size="0,10,5"

    )

    # loop to send chunk

    for i in range(chunk_num):



            beg = i * stride

            data = audio_bytes[beg:beg + stride]

 
            text = rcg.feed_chunk(data,wait_time=0.02)

            if len(text)>0:

               print("text",text)

            time.sleep(0.05)

 
        beg = i * stride

        data = audio_bytes[beg : beg + stride]



        text = rcg.feed_chunk(data, wait_time=0.02)

        if len(text) > 0:

            print("text", text)

        time.sleep(0.05)



    # get last message

    text = rcg.close(timeout=3)

    print("text",text)

 
    
            
    print("text", text)