devbox
繁依Fanyi0
这个屌丝很懒，什么也没留下！
关注作者
热门标签
热门文章
当前位置:   article > 正文

AI 文本语音互转的Python例子 (仅后台代码)_ai text voice

作者：繁依Fanyi0 | 2024-07-13 18:53:39

赞

踩

ai text voice

AI.py

  1. # -*- coding: utf-8 -*-
  2.  
  3. # ################## 入口类 ##################
  4. import os
  5. from flask import Flask,render_template,request
  6. from LogX import Log1
  7. from AI_TTS_Azure import Azure
  8. from AI_VoiceToText_Baidu import BaiduAi
  9.  
  10. log1 = Log1.log(None)
  11. # static_url_path='' 指明静态资源在 /static/ 下，URL 路径则可以直接省略 static 路径
  12. App3 = Flask(__name__, static_url_path='')
  13. App3.secret_key = os.urandom(32) #'U2FsdGVkX19B/bZVZXszkfQOBEseDukn' # AES,jiami,Hsbc123#
  14.  
  15. ########## 首页
  16. @App3.route('/', methods=['GET', 'POST'])
  17. @App3.route('/home', methods=['GET', 'POST'])
  18. @App3.route('/index', methods=['GET', 'POST'])
  19. def home():
  20.     return render_template('home.html')
  21.  
  22. ########## 文本转语音
  23. @App3.route('/index1', methods=['GET'])
  24. def index1():
  25.     return render_template('index1.html')
  26.  
  27. ########## 文本转语音
  28. @App3.route('/index1', methods=['POST'])
  29. def index1_tts():
  30.     result = Azure.ttsMain(None,request0=request)
  31.     return render_template('index1.html', msg=result['msg'], filepath=result['filepath'])
  32.  
  33. ########## 文本转语音
  34. @App3.route('/index2', methods=['GET'])
  35. def index2():
  36.     return render_template('index2.html')
  37.  
  38. ########## 文本转语音
  39. @App3.route('/index2', methods=['POST'])
  40. def index2_stt():
  41.     result = BaiduAi.sttMain(None,request0=request)
  42.     return render_template('index2.html', msg=result['msg'], filepath=result['filepath'])
  43.  
  44.  
  45. if __name__=='__main__':
  46.     App3.run(host='0.0.0.0', port=6789)
  47.  
  48.  
  49.  
  50.

AI_TTS_Azure.py

  1. # -*- coding: utf-8 -*-
  2.  
  3.  
  4. ### 文本转语音 Azure ###
  5. import os
  6. import requests
  7. import time
  8. from xml.etree import ElementTree
  9. from LogX import Log1
  10. from ConfigFile import Config
  11.  
  12. log1 = Log1.log(None)
  13. msg1 = '处理失败.'
  14. filepath1 = ''
  15.  
  16. '''
  17. 快速入门：https://docs.microsoft.com/zh-cn/azure/cognitive-services/speech-service/quickstart-python-text-to-speech
  18. 官方源码：https://github.com/Azure-Samples/Cognitive-Speech-TTS/blob/master/Samples-Http/Python/TTSSample.py
  19. 获取密钥：https://azure.microsoft.com/zh-cn/try/cognitive-services/my-apis/?apiSlug=speech-services&country=China&allowContact=true&fromLogin=True
  20. 终结点: https://westus.api.cognitive.microsoft.com/sts/v1.0
  21. 获取Token接口：https://westus.api.cognitive.microsoft.com/sts/v1.0/issueToken
  22. 文本转语音接口：https://westus.tts.speech.microsoft.com/cognitiveservices/v1
  23. 密钥 1: 8c0779e6846c43478f83fb2bd2fdd610
  24. 密钥 2: c7fa26e8ea4c46368ddafbe76fdf0827
  25. '''
  26.  
  27. proxyHttp   = Config.getBasic(None, 'http-proxy','http')
  28. proxyHttps  = Config.getBasic(None, 'http-proxy','https')
  29. azureKey1   = Config.getBasic(None, 'Azure-Key','key1')
  30. azureKey2   = Config.getBasic(None, 'Azure-Key','key2')
  31. azureKey    = azureKey1 if azureKey1=='' else azureKey2
  32.  
  33.  
  34. class Azure(object):
  35.  
  36.     def ttsMain(self,request0):
  37.         log1.warn(' >>> 进入 ttsMain ')
  38.         global msg1
  39.         global filepath1
  40.         nowtime1 = time.strftime('%Y%m%d%H%M%S')
  41.         texttype = request0.form['texttype']
  42.         caiyanglv = request0.form['caiyanglv']
  43.         yinse = request0.form['yinse']
  44.         voicefilename1 = 'voice_' + nowtime1 + '_[' + caiyanglv + ']_[' + yinse + '].wav'
  45.  
  46.         voice_key = ''
  47.  
  48.         if texttype != '':  # 文本类型<表单输入文本，上传文件文本，指定网页文本>
  49.             textcontent = ''
  50.             uploadPath = os.getcwd() + '/upload/'
  51.             if not os.path.exists(uploadPath):
  52.                 os.makedirs(uploadPath)
  53.  
  54.             #####[1] 表单文本
  55.             if texttype == 'formtext':
  56.                 textcontent = request0.form['text1']
  57.                 if textcontent != '':
  58.                     # log1.info(' 待转换的文本内容：\n' + textcontent)
  59.                     log1.warn(' 开始转换语音-表单文本 ...')
  60.                     filepath1 = Azure.textToVoice(None, text0=textcontent, voicefilename=voicefilename1,
  61.                                                   voice_encode_type=caiyanglv, voice_speeker_type=yinse,
  62.                                                   subscription_key=voice_key)
  63.                     if filepath1 != '':
  64.                         msg1 = '处理成功！'
  65.                     else:
  66.                         error = '处理失败！'
  67.                 else:
  68.                     msg1 = '处理失败！请输入文本内容'
  69.                     log1.error(' ' + msg1)
  70.  
  71.             #####[2] 文件文本
  72.             elif texttype == 'filetext':
  73.                 fileObj1 = request0.files['file1']
  74.                 if fileObj1 != '':
  75.                     filename0 = fileObj1.filename
  76.                     filename0_ext = filename0.split('.')[-1]
  77.                     filename0_ext_accept = ['txt', 'log', 'ini', 'conf']
  78.                     if filename0_ext not in filename0_ext_accept:
  79.                         msg1 = '处理失败！请上传纯文本格式的文件，支持的后缀：txt,log,ini,conf'
  80.                         log1.error(' ' + msg1)
  81.                     else:
  82.                         filenamenew = nowtime1 + '.' + filename0_ext
  83.                         abPath = uploadPath + '/text_' + filenamenew
  84.                         log1.info(' 开始上传文件 ...')
  85.                         fileObj1.save(abPath)
  86.                         log1.info(' 文件上传成功:' + abPath)
  87.                         fileUpload = open(abPath, 'r', encoding='utf-8').read()
  88.                         for textline in fileUpload:
  89.                             textcontent = textcontent + textline
  90.                         log1.info(' 开始转换语音-上传文件 ...')
  91.                         filepath1 = Azure.fileToVoice(None, filePath0=abPath, voicefilename=voicefilename1,
  92.                                                       voice_encode_type=caiyanglv, voice_speeker_type=yinse,
  93.                                                       subscription_key=voice_key)
  94.                         if filepath1 != '':
  95.                             msg1 = '处理成功！'
  96.                         else:
  97.                             msg1 = '处理失败！'
  98.                 else:
  99.                     msg1 = '处理失败！请选择文件'
  100.                     log1.error('>>> ' + msg1)
  101.  
  102.             #####[3] 网页文本
  103.             elif texttype == 'webpagetext':
  104.                 msg1 = '您好，暂不支持网页方式.'
  105.                 log1.info(' ' + msg1)
  106.                 '''
  107.                 weburl1 = request.form['weburl1']
  108.                 if textcontent != '':
  109.                     textcontent = weburl1 # 通过URL获取网页文本，应该加点过滤参数，获取指定DOM元素内的文本
  110.                     log1.info('>>> 开始转换语音-网页地址 ...')
  111.                     filepath1 = Azure.tts_text(None, text0=textcontent, voicefilename=voicefilename1,
  112.                                                voice_encode_type=caiyanglv, voice_speeker_type=yinse, subscription_key=voice_key)
  113.                     msg1 = '处理成功'
  114.                 else:
  115.                     msg1 = '处理失败！请输入网页地址(URL)'
  116.                     log1.info('>>> ' + msg1)
  117.                 '''
  118.             #####[0] 参数错误
  119.             else:
  120.                 msg1 = '处理失败！文本类型参数错误：texttype=' + texttype
  121.                 log1.error('>>> ' + msg1)
  122.  
  123.         else:
  124.             msg1 = '处理失败！文本类型参数不能为空.'
  125.             log1.error('>>> ' + msg1)
  126.  
  127.         if filepath1!='' :
  128.             filepath1 = request0.url_root + filepath1.replace('static/', '')
  129.  
  130.         result = {'msg': msg1, 'filepath': filepath1}  # 返回结果，键值对，字典
  131.         log1.warn('>>> 全部操作已经完成')
  132.         log1.warn('>>> 返回结果：\nresult=' + str(result))
  133.  
  134.         return result
  135.  
  136.  
  137.     ### 文件转语音
  138.     def fileToVoice(self, filePath0, voicefilename, voice_encode_type, voice_speeker_type, subscription_key):
  139.         filepath = ''
  140.  
  141.         log1.debug('传入的参数值：'
  142.                       + '\nfilePath0=' + filePath0
  143.                       + '\nvoicefilename=' + voicefilename
  144.                       + '\nvoice_encode_type=' + voice_encode_type
  145.                       + '\nvoice_speeker_type=' + voice_speeker_type)
  146.  
  147.         if subscription_key == '':
  148.             subscription_key = 'c7fa26e8ea4c46368ddafbe76fdf0827'
  149.         # tts = input('Input some text to convert to speech: ')  # 控制台输入
  150.         # tts = 'Hello everyone, I am 18 years old and have a deposit of 1.23 million. I want to get a wife.'
  151.  
  152.         file_obj = open(filePath0, 'rb') # 英文内容：'rU'，中文内容：,'rb'
  153.         # tts = ''
  154.         # for eachline in file_obj: # 一行一行读取
  155.         #     if eachline != '':
  156.         #         tts += eachline
  157.         # tts = file_obj.read() # 英文内容
  158.         tts = str(file_obj.read(), encoding='utf-8') # 中文内容
  159.         #log1.debug(' 待转换文本：' + tts )
  160.  
  161.         timestr = time.strftime('%Y%m%d%H%M')
  162.         proxies = {
  163.             'http': proxyHttp,
  164.             'https': proxyHttps,
  165.         }
  166.         fetch_token_url = 'https://westus.api.cognitive.microsoft.com/sts/v1.0/issueToken'
  167.         headers = {
  168.             'Ocp-Apim-Subscription-Key': subscription_key
  169.         }
  170.         response = requests.post(fetch_token_url, headers=headers, proxies=proxies, verify=False)
  171.         access_token = str(response.text)
  172.         #log1.debug(' 获取到Token：' + access_token)
  173.         constructed_url = 'https://westus.tts.speech.microsoft.com/cognitiveservices/v1'
  174.         headers = {
  175.             # 前面带有单词 Bearer 的授权令牌
  176.             'Authorization': 'Bearer ' + access_token,
  177.  
  178.             # 指定所提供的文本的内容类型。 接受的值：application/ssml+xml。
  179.             'Content-Type': 'application/ssml+xml',
  180.  
  181.             # 指定音频输出格式，取值如下：
  182.             # raw-16khz-16bit-mono-pcm  （百度AI的语音转文本时，需要采用这个格式，但Windows自带的播放器是无法播放的）
  183.             # raw-8khz-8bit-mono-mulaw
  184.             # riff-8khz-8bit-mono-alaw
  185.             # riff-8khz-8bit-mono-mulaw
  186.             # riff-16khz-16bit-mono-pcm
  187.             # audio-16khz-128kbitrate-mono-mp3
  188.             # audio-16khz-64kbitrate-mono-mp3
  189.             # audio-16khz-32kbitrate-mono-mp3
  190.             # raw-24khz-16bit-mono-pcm
  191.             # riff-24khz-16bit-mono-pcm （这个格式 可以播放）
  192.             # audio-24khz-160kbitrate-mono-mp3
  193.             # audio-24khz-96kbitrate-mono-mp3
  194.             # audio-24khz-48kbitrate-mono-mp3
  195.             'X-Microsoft-OutputFormat': voice_encode_type,
  196.  
  197.             # 应用程序名称，少于 255 个字符。
  198.             # Chrome的 User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/73.0.3683.86 Safari/537.36
  199.             'User-Agent': 'Chrome/73.0.3683.86'
  200.         }
  201.         xml_body = ElementTree.Element('speak', version='1.0')
  202.         xml_body.set('{http://www.w3.org/XML/1998/namespace}lang', 'en-us')
  203.         voice = ElementTree.SubElement(xml_body, 'voice')
  204.         voice.set('{http://www.w3.org/XML/1998/namespace}lang', 'en-US')
  205.         # 标准语音列表：https://docs.microsoft.com/zh-cn/azure/cognitive-services/speech-service/language-support#standard-voices
  206.         # 【下面5行：英语(美国)】	【语音性别，语音全称，语音简称】
  207.         # 女	'Microsoft Server Speech Text to Speech Voice (en-US, ZiraRUS)'	'en-US-ZiraRUS'
  208.         # 女	'Microsoft Server Speech Text to Speech Voice (en-US, JessaRUS)'	'en-US-JessaRUS'
  209.         # 男	'Microsoft Server Speech Text to Speech Voice (en-US, BenjaminRUS)'	'en-US-BenjaminRUS'
  210.         # 女	'Microsoft Server Speech Text to Speech Voice (en-US, Jessa24kRUS)'	'en-US-Jessa24kRUS'
  211.         # 男	'Microsoft Server Speech Text to Speech Voice (en-US, Guy24kRUS)'	'en-US-Guy24kRUS'
  212.         # 【下面3行：中文(中国)】
  213.         # 女	'Microsoft Server Speech Text to Speech Voice (zh-CN, HuihuiRUS)'	'zh-CN-HuihuiRUS'
  214.         # 女	'Microsoft Server Speech Text to Speech Voice (zh-CN, Yaoyao, Apollo)'	'zh-CN-Yaoyao-Apollo'
  215.         # 男	'Microsoft Server Speech Text to Speech Voice (zh-CN, Kangkang, Apollo)'	'zh-CN-Kangkang-Apollo'
  216.         voice.set('name',voice_speeker_type)
  217.         voice.text = tts
  218.         body = ElementTree.tostring(xml_body)
  219.         log1.debug(' 调用接口转换语音中......')
  220.         response = requests.post(constructed_url, headers=headers, data=body, proxies=proxies, verify=False)
  221.         if response.status_code == 200:
  222.             filepath = 'static/voice/' + voicefilename
  223.             with open(filepath, 'wb') as audio:
  224.                 audio.write(response.content)
  225.                 log1.warn(' 文本转语音已完成，生成的音频文件：' + filepath)
  226.         else:
  227.             print('[失败] response code: ' + str(response.status_code)
  228.                     + '\nresponse headers: ' + str(response.headers) )
  229.  
  230.         return filepath
  231.  
  232.     ### 文本转语音
  233.     def textToVoice(self, text0, voicefilename, voice_encode_type, voice_speeker_type, subscription_key):
  234.         filepath = ''
  235.  
  236.         log1.debug('传入的参数值：'
  237.                       #+ '\ntext0=' + text0
  238.                       + '\nvoicefilename=' + voicefilename
  239.                       + '\nvoice_encode_type=' + voice_encode_type
  240.                       + '\nvoice_speeker_type=' + voice_speeker_type )
  241.  
  242.         if subscription_key=='':
  243.             subscription_key = 'c7fa26e8ea4c46368ddafbe76fdf0827'
  244.         # tts = input('Input some text to convert to speech: ')  # 控制台输入
  245.         tts = text0
  246.         timestr = time.strftime('%Y%m%d%H%M')
  247.         proxies = {
  248.             'http': proxyHttp,
  249.             'https': proxyHttps,
  250.         }
  251.         fetch_token_url = 'https://westus.api.cognitive.microsoft.com/sts/v1.0/issueToken'
  252.         headers = {
  253.             'Ocp-Apim-Subscription-Key': subscription_key
  254.         }
  255.         response = requests.post(fetch_token_url, headers=headers, proxies=proxies, verify=False)
  256.         access_token = str(response.text)
  257.         #log1.debug(' 获取到Token：' + access_token)
  258.         constructed_url = 'https://westus.tts.speech.microsoft.com/cognitiveservices/v1'
  259.         headers = {
  260.             # 前面带有单词 Bearer 的授权令牌
  261.             'Authorization': 'Bearer ' + access_token,
  262.  
  263.             # 指定所提供的文本的内容类型
  264.             'Content-Type': 'application/ssml+xml',
  265.  
  266.             # 指定音频输出格式
  267.             'X-Microsoft-OutputFormat': voice_encode_type,
  268.  
  269.             # 应用程序名称，少于 255 个字符。
  270.             'User-Agent': 'Chrome/73.0.3683.86'
  271.         }
  272.         xml_body = ElementTree.Element('speak', version='1.0')
  273.         xml_body.set('{http://www.w3.org/XML/1998/namespace}lang', 'en-us')
  274.         voice = ElementTree.SubElement(xml_body, 'voice')
  275.         voice.set('{http://www.w3.org/XML/1998/namespace}lang', 'en-US')
  276.         # 'en-US-Guy24kRUS'，全称：'Microsoft Server Speech Text to Speech Voice (en-US, Guy24KRUS)'
  277.         voice.set('name',voice_speeker_type)
  278.         voice.text = tts
  279.         body = ElementTree.tostring(xml_body)
  280.         log1.debug(' 调用接口转换语音中......')
  281.         response = requests.post(constructed_url, headers=headers, data=body, proxies=proxies, verify=False)
  282.         if response.status_code == 200:
  283.             filepath = 'static/voice/' + voicefilename
  284.             with open(filepath, 'wb') as audio:
  285.                 audio.write(response.content)
  286.                 log1.warn(' 文本转语音已完成，生成的音频文件：' + filepath)
  287.         else:
  288.             print('[失败] response code: ' + str(response.status_code)
  289.                     + '\nresponse : ' + str(response.content) )
  290.         return filepath
  291.  
  292. ### 函数入口
  293. if __name__ == '__main__':
  294.     test = Azure()
  295.     abPath = 'static/backup/text/tts_txt_zhcn.txt'
  296.     caiyanglv = 'riff-24khz-16bit-mono-pcm'
  297.     yinse = 'zh-CN-HuihuiRUS'
  298.     voicefilename1 = 'voice_1001_[' + caiyanglv + ']_['+yinse+'].wav'
  299.     test.fileToVoice( filePath0 = abPath, voicefilename = voicefilename1, voice_encode_type = caiyanglv,
  300.              voice_speeker_type = yinse, subscription_key = '')
  301.

AI_VoiceToText_Baidu.py

  1. # -*- coding: utf-8 -*-
  2.  
  3.  
  4. ### 语音识别 百度API ###
  5. # Home:https://ai.baidu.com/tech/speech/asr
  6. # API: https://ai.baidu.com/docs#/ASR-Online-Python-SDK/top
  7. # Q&A: https://ai.baidu.com/docs#/FAQ/a53b4698
  8. # 在线文本转语音（汇丰无法试听和下载）：https://developer.baidu.com/vcast
  9. from aip import AipSpeech
  10. from ConfigFile import Config
  11. from LogX import Log1
  12. import time
  13. import os
  14.  
  15. log1 = Log1.log(None)
  16. msg1 = '处理失败.'
  17. filepath1 = ''
  18.  
  19.  
  20. class BaiduAi(object):
  21.  
  22.     """ 你的 APPID AK SK """
  23.     proxyHttp = Config.getBasic(None, group0='http-proxy', key0='http')
  24.     proxyHttps = Config.getBasic(None, group0='http-proxy', key0='https')
  25.     APP_ID = Config.getBasic(None, group0='Baidu-Ai', key0='appid')
  26.     API_KEY = Config.getBasic(None, group0='Baidu-Ai', key0='appkey')
  27.     SECRET_KEY = Config.getBasic(None, group0='Baidu-Ai', key0='secretkey')
  28.  
  29.     # ### 语音转文本，表单入口
  30.     def sttMain(self, request0):
  31.         global msg1
  32.         global filepath1
  33.         nowtime1 = time.strftime('%Y%m%d%H%M%S')
  34.         fromLanguage = request0.form['fromLanguage'] # 来源语言
  35.         toLanguage   = request0.form['toLanguage']   # 目标语言
  36.         textfilename1 = 'text_' + nowtime1 + '_[' + fromLanguage + ']_[' + toLanguage + '].txt'
  37.         uploadPath = os.getcwd() + '/upload/'
  38.         fileObj1 = request0.files['file1']
  39.         if fileObj1 != '':
  40.             filename0 = fileObj1.filename
  41.             filename0_ext = filename0.split('.')[-1]
  42.             filename0_ext_accept = ['wav']
  43.             if filename0_ext not in filename0_ext_accept:
  44.                 msg1 = '处理失败！请上传wav格式的音频文件。'
  45.                 log1.error('>>> ' + msg1)
  46.             else:
  47.                 filenamenew = nowtime1 + '.' + filename0_ext
  48.                 abPath = uploadPath + '/voice_' + filenamenew
  49.                 log1.info('>>> 开始上传文件 ...')
  50.                 fileObj1.save(abPath)
  51.                 log1.info('>>> 文件上传成功:' + abPath)
  52.  
  53.                 # 开始转换语音
  54.                 textDict = BaiduAi.aiMain(None, filePath0=abPath, toLanguage0=toLanguage)
  55.                 if textDict != '':
  56.                     err_no  = textDict['err_no']
  57.                     err_msg = textDict['err_msg']
  58.                     result  = textDict['result'][0]
  59.                     log1.debug('>>> 状态码：' + str(err_no))
  60.                     log1.debug('>>> 状态：' + err_msg)
  61.                     log1.debug('>>> 结果：' + result)
  62.  
  63.                     # TODO 将文本写入文件
  64.                     filepath1 = 'static/textfile/'+textfilename1
  65.                     open(filepath1, "w", encoding='UTF-8').write(result)
  66.                     log1.debug('>>> 写入文件完成：' + filepath1)
  67.                     filepath1 = request0.url_root + 'textfile/' + textfilename1
  68.  
  69.                 if filepath1 != '':
  70.                     msg1 = '处理成功！'
  71.                 else:
  72.                     msg1 = '处理失败！'
  73.         else:
  74.             msg1 = '处理失败！请选择文件'
  75.             log1.error('>>> ' + msg1)
  76.  
  77.         result = {'msg': msg1, 'filepath': filepath1}  # 返回结果，键值对，字典
  78.         log1.warn('>>> 全部操作已经完成')
  79.         log1.warn('>>> 返回结果：\nresult=' + str(result))
  80.  
  81.         return result
  82.  
  83.     # ### 语音转文本，转换函数，返回字典类型
  84.     def aiMain(self, filePath0, toLanguage0):
  85.  
  86.         # AipSpeech是语音识别的Python SDK客户端
  87.         client = AipSpeech(BaiduAi.APP_ID, BaiduAi.API_KEY, BaiduAi.SECRET_KEY)  # 访问百度API，内置访问http的函数，并写死了URL路径
  88.         proxies = {
  89.             "http": BaiduAi.proxyHttp,
  90.             "https": BaiduAi.proxyHttps,
  91.         }
  92.         client.setProxies(proxies)
  93.         # client.setConnectionTimeoutInMillis( 10000 )  # 建立连接的超时时间（毫秒），默认不需要设置
  94.         # client.setSocketTimeoutInMillis(300000)  # 通过打开的连接传输数据的超时时间（毫秒）
  95.  
  96.         # 上传 本地音频文件（SpeechToText.wav）
  97.         # 支持的格式：原始 PCM 的录音参数必须符合 16k 采样率、16bit 位深、单声道，支持的格式有：pcm（不压缩）、wav（不压缩，pcm编码）、amr（压缩格式）。
  98.         # 有长度限制（测试时尽量短一点）
  99.         # 返回结果为字典类型（Dict / json）
  100.         response = client.asr(BaiduAi.get_file_content(None,filePath=filePath0), 'wav', 16000, {
  101.             # 1536：普通话(支持简单的英文识别)
  102.             # 默认1537：普通话(纯中文识别)
  103.             # 1737：英语
  104.             # 1637：粤语
  105.             'dev_pid': int(toLanguage0),
  106.         })
  107.  
  108.         return response
  109.         # {'corpus_no': '6714166347641437907', 'err_msg': 'success.', 'err_no': 0, 'result': ['识别后的内容'], 'sn': '602626830571563263672'}
  110.  
  111.     # ### 语音转文本，读取本地文件
  112.     def get_file_content(self,filePath):
  113.         with open(filePath, 'rb') as fp:
  114.             return fp.read()
  115.  
  116. ### 函数入口
  117. if __name__ == "__main__":
  118.     baidu = BaiduAi()
  119.     filepath9 = 'static/backup/sound/voice_20190916142927_[raw-16khz-16bit-mono-pcm]_[en-US-ZiraRUS].wav'
  120.     languageNum = 1737
  121.     textDict = baidu.aiMain( filePath0=filepath9, toLanguage0=languageNum)
  122.     print (str(textDict))
  123.     if textDict!='':
  124.         err_no  = textDict['err_no']
  125.         err_msg = textDict['err_msg']
  126.         result  = textDict['result'][0]
  127.         print ('状态码：' + err_no)
  128.         print ('状态：' + err_msg)
  129.         print ('结果：' + result)
  130.  
  131.

ConfigFile

  1. # -*- coding: utf-8 -*-
  2.  
  3. import os
  4. from configparser import ConfigParser
  5.  
  6.  
  7. class Config(object):
  8.  
  9.     def __init__(self):
  10.         super(Config, self).__init__()
  11.         self.config = ConfigParser()
  12.         #print('配置文件路径：'+os.getcwd()+'/configs/basic.ini')
  13.         self.config.read(os.getcwd()+'/configs/basic.ini',encoding='utf-8')
  14.  
  15.     # ### 代理配置
  16.     # user
  17.     # pass
  18.     def getBasic (self,group0, key0):
  19.         cfg0    = Config().config
  20.         value0  = cfg0.get(group0, key0)
  21.         return value0
  22.  
  23. if __name__ == "__main__":
  24.     cc = Config()
  25.     # print (cc.getBasic( group0='http-proxy', key0='http' ))
'
运行

LogX.py

  1. # -*- coding: utf-8 -*-
  2.  
  3.  
  4. import logging
  5. import logging.handlers
  6.  
  7. class Log1(object):
  8.  
  9.     def log(self):
  10.         # 日志对象
  11.         logger1 = logging.getLogger('AIFuture')
  12.         if not logger1.handlers:
  13.             # 日志级别-会覆盖 Handler的子级别
  14.             logfilePath = 'logs/AI.log'
  15.             logger1.setLevel(logging.DEBUG)
  16.             # 用于输出到文件
  17.             #logfile1 = logging.FileHandler(logfilePath)
  18.             logfile1 = logging.handlers.RotatingFileHandler(logfilePath, maxBytes=10 * 1024 * 1024, backupCount=5, encoding='utf-8')
  19.             #logfile1.setLevel(logging.DEBUG)
  20.             # 用于输出到控制台
  21.             logcmd1 = logging.StreamHandler()
  22.             #logcmd1.setLevel(logging.INFO)
  23.             # 格式化日志
  24.             logformatter = logging.Formatter('[%(asctime)s] [%(levelname)s] [%(filename)s(line:%(lineno)d)] %(message)s')
  25.             # 应用格式
  26.             logfile1.setFormatter(logformatter)
  27.             logcmd1.setFormatter(logformatter)
  28.             # 注入对象
  29.             logger1.addHandler(logcmd1)
  30.             logger1.addHandler(logfile1)
  31.  
  32.         return logger1
'
运行

start_AI.sh

  1. #! /bin/bash
  2. . ~/.bash_profile
  3. mypython -u AI.py
  4.  
  5.  
  6. '''
  7. 进入目录： cd /home/pyai/tts
  8. 启动命令： nohup sh start_AI.sh > start_AI.sh.out &
  10. 停止命令：
  11. 查看进程： ps -ef|grep AI
  12. 显示 端口占用：netstat -lnt|grep 6789
  14. sudo netstat -antup
  15. sudo netstat -antup|grep PID号 
  16. sudo ps -ef|grep 进程名
  18. '''
  19.  
  20. '''
  21. 依赖插件：mypip install configparser --user
  22. flask
  23. jinja2
  24. requests
  25. logger1
  26. baidu-aip
  27. configparser
  28. '''
  29.

configs/basic.ini

  1. # http 代理配置
  2. [http-proxy]
  3. http  = http://***
  4. https = http://***
  5.  
  6.  
  7. # 微软 Azure 语音密钥
  8. [Azure-Key]
  9. key1 = 8c0779e6846c43478f83fb2bd2fdd610
  10. key2 = c7fa26e8ea4c46368ddafbe76fdf0827
  11.  
  12. [Baidu-Ai]
  13. appid     = 16735297
  14. appkey    = P7IGlBm443VcSaCWCQT6WrXh
  15. secretkey = pdlslL0ysPFUU8yF6dQtoN5yxrfdSupD

相关目录结构：

  1. 项目根目录/configs/basic.ini
  2.  
  3. 项目根目录/logs
  4.  
  5. 项目根目录/static/backup/sound
  6. 项目根目录/static/backup/text
  7.  
  8. 项目根目录/static/history
  9.  
  10. 项目根目录/static/textfile
  11. 项目根目录/static/voice
  12.  
  13. 项目根目录/templates/***.html
  14. 项目根目录/upload
  15. 项目根目录/AI.py
  16. 项目根目录/AI_TTS_Azure.py
  17. 项目根目录/AI_VoiceToText_Baidu.py
  18. 项目根目录/ConfigFile.py
  19. 项目根目录/LogX.py
  20. 项目根目录/start_AI.py

声明：本文内容由网友自发贡献，不代表【wpsshop博客】立场，版权归原作者所有，本站不承担相应法律责任。如您发现有侵权的内容，请联系我们。转载请注明出处：https://www.wpsshop.cn/w/繁依Fanyi0/article/detail/821058
推荐阅读
相关标签

Copyright © 2003-2013 www.wpsshop.cn 版权所有，并保留所有权利。

  

闽ICP备14008679号