代碼如下：

# -*- coding:utf-8 -*- #http://blog.csdn.net/happen23/article/details/45821697 #百度語音識(shí)別API的使用樣例（python實(shí)現(xiàn)） #encoding=utf-8import wave import urllib, urllib2, pycurl import base64 import json ## get access token by api key & secret keydef get_token():apiKey = "xjlSpsvUgGF8a9ltNOtREoTr"secretKey = "a95ca71b81854b526e7eb04ae8f51d23"auth_url = "https://openapi.baidu.com/oauth/2.0/token?grant_type=client_credentials&client_id=" + apiKey + "&client_secret=" + secretKey;res = urllib2.urlopen(auth_url)json_data = res.read()return json.loads(json_data)['access_token']def dump_res(buf):print buf## post audio to server def use_cloud(token):fp = wave.open('8k.wav', 'rb')nf = fp.getnframes()f_len = nf * 2audio_data = fp.readframes(nf)cuid = "xxxxxxxxxx" #my xiaomi phone MACsrv_url = 'http://vop.baidu.com/server_api' + '?cuid=' + cuid + '&token=' + tokenhttp_header = ['Content-Type: audio/pcm; rate=8000','Content-Length: %d' % f_len]c = pycurl.Curl()c.setopt(pycurl.URL, str(srv_url)) #curl doesn't support unicode#c.setopt(c.RETURNTRANSFER, 1)c.setopt(c.HTTPHEADER, http_header) #must be list, not dictc.setopt(c.POST, 1)c.setopt(c.CONNECTTIMEOUT, 30)c.setopt(c.TIMEOUT, 30)c.setopt(c.WRITEFUNCTION, dump_res)c.setopt(c.POSTFIELDS, audio_data)c.setopt(c.POSTFIELDSIZE, f_len)c.perform() #pycurl.perform() has no return valif __name__ == "__main__":token = get_token()use_cloud(token)

以上代碼是不能直接跑的，一般錄音得到的音頻文件，該代碼下運(yùn)行不了，目前只知道百度提供的音頻文件可以識(shí)別，百度提供的音頻文件下載鏈接如下：

http://yuyin.baidu.com/docs/asr/54

在上面這個(gè)鏈接的頁(yè)面中，往下拖，可以得到下載鏈接如下：
http://speech-doc.gz.bcebos.com/rest-api-asr/public_audio/public.zip

然后運(yùn)行結(jié)果：

{"corpus_no":"6485972281376050071","err_msg":"success.","err_no":0,"result":["北京科技館，"],"sn":"651708407021510133099"}

注意事項(xiàng)：

百度的語音識(shí)別和語音合成用的是相同的

appid、API key和Secret Key，所以使用相同的token即可

獲取以上三個(gè)字段的教程：

http://jingyan.baidu.com/article/f3e34a12df0cddf5eb65359f.html

后記:

下面的代碼可以運(yùn)行任意自己錄制的音頻文件，注意，運(yùn)行前必須apt-get install ffmpeg

另外，rate改成了16000，不然會(huì)識(shí)別不準(zhǔn)，不過，也沒有經(jīng)過大量測(cè)試，不知道識(shí)別準(zhǔn)確的情況還會(huì)不會(huì)出現(xiàn)。

# -*- coding:utf-8 -*- #http://blog.csdn.net/happen23/article/details/45821697 #百度語音識(shí)別API的使用樣例（python實(shí)現(xiàn)） #encoding=utf-8import wave import urllib, urllib2, pycurl import base64 import subprocess import json ## get access token by api key & secret keydef get_token():apiKey = "xjlSpsvUgGF8a9ltNOtREoTr"secretKey = "a95ca71b81854b526e7eb04ae8f51d23"auth_url = "https://openapi.baidu.com/oauth/2.0/token?grant_type=client_credentials&client_id=" + apiKey + "&client_secret=" + secretKey;res = urllib2.urlopen(auth_url)json_data = res.read()return json.loads(json_data)['access_token']def dump_res(buf):print buf## post audio to server def use_cloud(token):subprocess.call(['ffmpeg', '-i', 'tian.mp3', 'tian.wav'])#這句代碼的意思是在終端中執(zhí)行[]中的命令。所以執(zhí)行的前提是apt-get install ffmpegfp = wave.open('tian.wav', 'rb')nf = fp.getnframes()f_len = nf * 2audio_data = fp.readframes(nf)cuid = "xxxxxxxxxx" #my xiaomi phone MACsrv_url = 'http://vop.baidu.com/server_api' + '?cuid=' + cuid + '&token=' + tokenhttp_header = ['Content-Type: audio/pcm; rate=16000','Content-Length: %d' % f_len]c = pycurl.Curl()c.setopt(pycurl.URL, str(srv_url)) #curl doesn't support unicode#c.setopt(c.RETURNTRANSFER, 1)c.setopt(c.HTTPHEADER, http_header) #must be list, not dictc.setopt(c.POST, 1)c.setopt(c.CONNECTTIMEOUT, 30)c.setopt(c.TIMEOUT, 30)c.setopt(c.WRITEFUNCTION, dump_res)c.setopt(c.POSTFIELDS, audio_data)c.setopt(c.POSTFIELDSIZE, f_len)c.perform() #pycurl.perform() has no return valif __name__ == "__main__":token = get_token()use_cloud(token)所謂的離線資源下載，其實(shí)仍然是本地向服務(wù)器請(qǐng)求，效率上是無法提高的。

創(chuàng)作挑戰(zhàn)賽新人創(chuàng)作獎(jiǎng)勵(lì)來咯，堅(jiān)持創(chuàng)作打卡瓜分現(xiàn)金大獎(jiǎng)

總結(jié)

以上是生活随笔為你收集整理的python使用百度语音识别API注意事项的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。