树莓派用百度语音的 API 做语音识别，识别率奇低

[greatdk]

最近用树莓派在做语音方面的东西玩，用到了百度语音的语音识别接口

我用的是REST_API，利用python来使用，但是不知道为啥，能调用成功，但是识别率奇低，不管说啥识别出来都是嗯，哈哈，之类的，我直接播放语音，觉得听的还是挺清楚的，百度语音貌似不会这么低啊，放上代码：

#encoding=utf-8

import wave
import urllib, urllib2, pycurl
import base64
import json
## get access token by api key & secret key

def get_token():
apiKey = "xxxxxxxx"
secretKey = "xxxxxxxxx"

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('vad_0.wav', 'rb')
nf = fp.getnframes()
f_len = nf * 2
audio_data = fp.readframes(nf)

cuid = "xxxxxxxxxx" #my xiaomi phone MAC
srv_url = 'http://vop.baidu.com/server_api' + '?cuid=' + cuid + '&token=' + token
http_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 dict
c.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 val

if __name__ == "__main__":
token = get_token()
use_cloud(token)

[nkssai]

换个音频格式试一试？ 之前用google语音接口的时候发现不同的格式识别准确度还有不小的差别。