失眠网 > 语音识别（利用python将语音转化为文字）（升级版）

语音识别（利用python将语音转化为文字）（升级版）

时间：2020-09-27 19:31:27

提示：文章写完后，目录可以自动生成，如何生成可参考右边的帮助文档

文章目录

前言一、申请讯飞语音端口1.点击链接进入讯飞平台主页面2.在页面注册自己的个人账户3.申请语音端口4.查看自己的端口编码二、python代码讲解1.代码如下（实例）2.代码需要修改的部分3.包的导入4.结果演示总结

前言

基于语音识别（1）进行的完善，修改了60秒断触的问题，另外可以更加方便的调用，语音识别1的链接如下：

/m0_46657126/article/details/124531081

一、申请讯飞语音端口

1.点击链接进入讯飞平台主页面

2.在页面注册自己的个人账户

ps：注册账户是完全免费的，因为我之前已经注册过一个了，这里就不重复介绍了，注册之后看个人情况进行个人认证，这个不影响之后的操作。

3.申请语音端口

ps:申请内容大概像我这样写就行，名字重复的话换一个

4.查看自己的端口编码

二、python代码讲解

1.代码如下（实例）

代码如下（示例）：

# -*- encoding:utf-8 -*-import hashlibimport hmacimport base64from socket import *import json, time, threadingfrom websocket import create_connectionimport websocketfrom urllib.parse import quoteimport loggingimport pyaudioimport re# reload(sys)# sys.setdefaultencoding("utf8")class Client():def __init__(self):logging.basicConfig()self.app_id = ""self.api_key = ""base_url = "ws:///v1/ws"ts = str(int(time.time()))tt = (self.app_id + ts).encode('utf-8')md5 = hashlib.md5()md5.update(tt)baseString = md5.hexdigest()baseString = bytes(baseString, encoding='utf-8')apiKey = self.api_key.encode('utf-8')signa = hmac.new(apiKey, baseString, hashlib.sha1).digest()signa = base64.b64encode(signa)signa = str(signa, 'utf-8')self.end_tag = "{\"end\": true}"self.ws = create_connection(base_url + "?appid=" + self.app_id + "&ts=" + ts + "&signa=" + quote(signa))self.trecv = threading.Thread(target=self.recv)self.trecv.start()def send(self):CHUNK = 300 # 定义数据流块FORMAT = pyaudio.paInt16 # 16bit编码格式CHANNELS = 1 # 单声道RATE = 16000 # 16000采样频率p = pyaudio.PyAudio()# 创建音频流stream = p.open(format=FORMAT, # 音频流wav格式channels=CHANNELS, # 单声道rate=RATE, # 采样率16000input=True,frames_per_buffer=CHUNK)print("- - - - - - - Start Recording ...- - - - - - - ")while True:# file_object = stream.read(CHUNK)index = 1while True:chunk = stream.read(1280)if not chunk:breakself.ws.send(chunk)index += 1time.sleep(0.04)# self.ws.send(bytes(self.end_tag.encode('utf-8')))# print("send end tag success")def recv(self):try:while self.ws.connected:result = str(self.ws.recv())if len(result) == 0:print("receive result end")breakresult_dict = json.loads(result)# 解析结果if result_dict["action"] == "started":print("handshake success, result: " + result)if result_dict["action"] == "result":result = ''result_1 = re.findall('"w":"(.*?)"', str(result_dict["data"]))for i in result_1:if i == '。' or i == '.。' or i == ' .。' or i == ' 。':passelse:result += iprint("翻译结果：" + result)# 写入文本文件tep = open('tep.txt', 'w', encoding='utf-8')tep.write(result)tep.close()# print("rtasr result: " + result_1)if result_dict["action"] == "error":print("rtasr error: " + result)self.ws.close()returnexcept websocket.WebSocketConnectionClosedException:print("receive result end")def close(self):self.ws.close()print("connection closed")def runc():client = Client()client.send()if __name__ == '__main__':logging.basicConfig()client = Client()client.send()