树莓派c语言百度语音识别,树莓派语音识别

8种机械键盘轴体对比

本人程序员，要买一个写代码的键盘，请问红轴和茶轴怎么选？

参考：

配置：

喇叭：3.5音频口

麦克风：摄像头自带麦克风(Usb)

1. 安装需要的软件1

2sudo apt-get install python-pip mplayer flac python2.7-dev libcurl4-gnutls-dev

sudo pip install requests pycurl

2. 配置配置麦克风和音箱1

2

3

4

5

6

7

8

9

10$ cat /proc/asound/cards

0 [ALSA ]: bcm2835 - bcm2835 ALSA

bcm2835 ALSA

1 [Camera ]: USB-Audio - USB 2.0 Camera

Sonix Technology Co., Ltd. USB 2.0 Camera at usb-3f980000.usb-1.2, high speed

$ cat /proc/asound/modules

0 snd_bcm2835

1 snd_usb_audio

第一个命令查看声卡是否存在

第二个命令应该是看声卡的驱动的

调整输入输出音频工具alsamixer：sudo alsamixer

3. 播放声音aplay make.wav

结果声音没有播放

网上查了，

By default output will be automatic (hdmi if hdmi supports audio, otherwise analogue). You can force it with:amixer cset numid=3 where n is 0=auto, 1=headphones, 2=hdmi.

If you have pulseaudio installed you need to also specify the card number:amixer -c 0 cset numid=3

所以运行amixer cset numid=3 1

就可以听到3.5输出了

4. 麦克风测试sudo arecord –duration=10 –device=plughw:1,0 –format=cd aaa.wav

指令中的’–device=plughw:1,0’ 选项指派特定的录音设备，而Raspberry Pi的内置音效设备是’plughw:0,0’

使用百度api进行语音播放：

5. 使用百度API进行语音识别和播放

测试取的token：1

2

3

4

5

6

7

8

9

10

11import requests

import json

url = 'https://openapi.baidu.com/oauth/2.0/token'

payload = {'grant_type':'client_credentials',

'client_id':'ZEzwlE2gipnqoE3jqiCKN7SN',

'client_secret':'KOyWkyE57qWnDnFoc5afwIk0zBMw2X1p'}

r = requests.post(url,data=payload)

print json.loads(r.text)['access_token']

测试ok后，开始正式写：

大致思路：

将请求的token保存到文件中，然后读取文件中的accesstoken,判断token的日期是否有效，无效重新请求token。

代码：

app.py1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131# -*- coding:utf8 -*-

import requests

import os

import json

import time

import uuid

import base64

class API(object):

def __init__(self):

dic = {

'hello':'欢迎使用语音播放',

'error':'识别错误',

}

for key in dic:

self.txt2audio(dic[key],key)

if key == 'hello':

self.play(key)

pass

def get_token(self):

token = None

if os.path.exists('token'):

file = open('token','r')

token = file.read()

file.close()

if json.loads(token)['expires'] < time.time():

token = self.req_token()

else:

print "load cache token"

pass

else:

token = self.req_token()

return json.loads(token)['access_token']

def req_token(self):

print "request new token"

token = None

url = 'https://openapi.baidu.com/oauth/2.0/token'

key = 'ZEzwlE2gipnqoE3jqiCKN7SN'

secret = 'KOyWkyE57qWnDnFoc5afwIk0zBMw2X1p'

payload = {'grant_type':'client_credentials',

'client_id':key,

'client_secret':secret}

r = requests.post(url,data=payload)

token = r.text

jtoken = json.loads(token)

jtoken['expires'] = jtoken['expires_in'] + time.time()

token = json.dumps(jtoken)

file = open('token','w')

file.write(token)

file.close()

return token

def txt2audio(self,txt,audio='tmp'):

access_token = self.get_token()

url = "http://tsn.baidu.com/text2audio"

payload = {

'tex':txt,

'lan':'zh',

'tok':access_token,

'ctp':1,

'cuid':uuid.UUID(int = uuid.getnode()).hex,

'spd':5,

'pit':8,

'vol':9,

'per':1,

}

r = requests.post(url,payload)

if r.headers['Content-Type'] == 'audio/mp3':

file = open(audio + '.mp3','w')

file.write(r.content)

file.close()

else:

self.play('error')

print r.text

pass

def audio2txt(self,file='rec'):

access_token = self.get_token()

url="http://vop.baidu.com/server_api"

self.txt2audio('请说:')

self.play('tmp')

os.system('arecord --duration=2 -c 1 --device=plughw:1,0 -f S16_LE -t wav -v --rate=16000 ' + file + '.wav')

f = open(file + '.wav','r')

sound = f.read()

f.close()

payload = {

'format':'wav',

'rate':16000,

'channel':1,

'cuid':uuid.UUID(int = uuid.getnode()).hex,

'token':access_token,

'lan':'zh',

'speech':base64.b64encode(sound),

'len':len(sound),

}

header = {'Content-Type':'application/json'}

r = requests.post(url,data=json.dumps(payload),headers=header)

print r.text

jret = json.loads(r.text)

if jret.has_key('result'):

ret = json.loads(r.text)['result'][0].split(',')[0]

else:

ret = ""

return ret

pass

def play(self,file='tmp'):

# can not play the mp3 by aplay

# os.system("aplay " + file + ".mp3")

os.system("mpg321 " + file + ".mp3")

pass

baidu.py1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

47

48# -*- coding:utf8 -*-

from baidu import API

import RPi.GPIO as GPIO

import time

import requests

import uuid

import json

pin = 13

GPIO.setmode(GPIO.BCM)

GPIO.setup(pin, GPIO.IN)

baidu = API()

# baidu.txt2audio("你好")

# baidu.play('tmp')

def get_tuling(text):

url='http://www.tuling123.com/openapi/api'

data={'key':'fa78fe2fbb85c914c7126d42bc7c3ebb','info':text,'userid':str(uuid.UUID(int = uuid.getnode()).hex,)}

r = requests.post(url,data=data)

ans = json.loads(r.text)

return ans

while True:

change = False

while GPIO.input(pin) == GPIO.LOW:

change = True

time.sleep(0.005)

if change == True:

#baidu.txt2audio('我听着,请说指令:')

text = baidu.audio2txt()

print text

if len(text) > 0:

rep = get_tuling(text)

rep = rep['text']

print rep

else:

rep = "我没有听清"

baidu.txt2audio(rep)

baidu.play()

change = False

time.sleep(1)

遇到坑：mp3格式只能通过mpg321播放

读取文件token = file.readlines()[0]，是一个列表，所以需要先取第一个再用json转。