def try_getstr():
pyrec.rec("1.wav")
pcm_file = wav2pcm.wav_to_pcm("1.wav")
# 识别本地文件
with open(pcm_file, "rb") as fp:
file_context = fp.read()
result = client.asr(file_context, 'pcm', 16000, {
'dev_pid': 1536,
})
print(result)
result = result.get('result', None)
if result:
print(result)
return result
def processing(limit, begin):
i = 0
print('begin', begin.value)
while True:
# 请说出语音指令
print("\n\n==================================================")
rec_yuyin = "D:/Python/self_project/luyin/files/test" + str(i) + ".wav"
#print(rec_yuyin)
luyin.audio_record(rec_yuyin, 3)
pcm_file = wav2pcm.wav_to_pcm(rec_yuyin)
asr_result = shibie.shibie(pcm_file) # 识别语音指令
if len(asr_result) != 0:
print("Identify Result:", asr_result)
print("Start Control...")
control.keyboard_control(asr_result, limit, begin) # 根据识别结果控制键盘
print("Control End...")
if asr_result.find("退出") != -1: # 如果是"退出"指令则结束程序
break
time.sleep(0.5) # 延时1秒
i = i + 1
def asr_main(audio_data_va):
pcm_file = wav2pcm.wav_to_pcm(audio_data_va)
signal = open(pcm_file, "rb").read()
token = get_access_token()
## token = "以上获取token令牌可以保持下来，不用一直获取，一个月有效"
uuid = "8C-16-45-44-09-E1" # 机器 MAC 地址
#print(resp)
try:
resp = baidu_asr(signal, uuid, token) # resp为list格式
if '，' not in resp:
res = ''.join(resp) # 将list格式转化为str
else:
res = "None"
except:
print("音频质量差")
res = "None"
print(res)
return res
def voice_recognize():
record.rec('test1.wav') #生成录音
time_start = time.time() #计时开始
pcm_file = wav2pcm.wav_to_pcm('test1.wav') #将wav转成pcm
# 识别本地文件
result = client.asr(get_file_content(pcm_file), 'pcm', 16000, {
'dev_pid': 1536,
})
if (result.get("err_no") == 0):
time_end = time.time() #计时结束
res_str = '识别结果：%s,耗时%.2fs' % (result.get("result")[0],
(time_end - time_start)) #截取结果
wav2pcm.deletef(pcm_file) #将此次识别的pcm文件删除
else:
res_str = "识别失败"
wav2pcm.deletef(pcm_file) #将此次识别的pcm文件删除
return res_str
def main():
while True:
client = AipSpeech(APP_ID, API_KEY, SECRET_KEY)
pyrec.rec("1.wav") # 录音并生成wav文件,使用方式传入文件名
pcm_file = wav2pcm.wav_to_pcm("1.wav") # 将wav文件 转换成pcm文件 返回 pcm的文件名
# 读取文件 , 得到了PCM文件
with open(pcm_file, 'rb') as fp:
file_context = fp.read()
# 识别本地文件
res = client.asr(file_context, 'pcm', 16000, {
'dev_pid': 1536,
})
# 从字典里面获取"result"的value 列表中第1个元素,就是识别出来的字符串
res_str = res.get("result")[0]
print(res_str)
#识别出来的内容交给图灵机器人来回答
Q_str = FAQ.answer(res_str, TULING_KEY)
# 将转换出来的字符串进行语音合成
synth_file = "synth.mp3"
synth_context = client.synthesis(Q_str, "zh", 1, {
"vol": 6,
"spd": 2,
"pit": 3,
"per": 0
})
with open(synth_file, "wb") as f:
f.write(synth_context)
# 播放合成语音
pygame.mixer.init()
pygame.mixer.music.load(synth_file)
pygame.mixer.music.play()
count = 0
while (count < 9):
csx = pygame.mixer.music.get_busy()
if csx == 1:
time.sleep(5)
count = count + 1
else:
count = 10
pygame.mixer.music.stop()
def parse_voice_file(filename):
cmd_say('开始解析语音文件')
newFilePath = wav_to_pcm(filename)
print("Handling file ", filename)
# 通过百度语音识别API识别本地录制并已转化的语音文件
resp = client.asr(get_file_content(newFilePath), 'pcm', 16000, {
'dev_pid': 1536,
})
print("AI response:", resp)
print("")
result = resp['result']
cmd_texts = []
for r in result:
print(r)
cmd_texts.append(r)
cmd_say(u"文件解析的结果是 {}".format(r))
for text in cmd_texts:
# We can implement some strategies here
if '音乐' in text:
cmd_say("检测到有'音乐'这个关键字，正在打开音乐...")
os.system("open -a NeteaseMusic")
time.sleep(2)
# 播放，需要设置全局快捷键
keyboard.press_and_release('ctrl+v')
else:
if '浏览器' in text:
cmd_say("检测到有'浏览器'这个关键字，正在打开Chrome...")
os.system("open -a Google\ Chrome")
time.sleep(2)
def audio2text():
APP_ID = '11567018'
API_KEY = 'v8P6XA1I4A2EGnPRQmpt9D99'
SECRET_KEY = 'llbHhdFrRe2TFugk2RbNyCdx6yjLMjAq'
client = AipSpeech(APP_ID, API_KEY, SECRET_KEY)
pyrec.rec("1.wav")
pcm = wav2pcm.wav_to_pcm("1.wav")
with open('1.pcm', 'rb') as fp:
file_context = fp.read()
data = client.asr(file_context, 'pcm', 16000, {'dev_pid': 1536})
if (data[u'err_msg'] == u'success.'):
with open('test.txt','w+') as f:
for i in data[u'result']:
print i
f.write(i.encode("utf-8"))
return True
else:
return False
#!/usr/bin/python
# -*- coding: utf-8 -*-
import pyrec
import wav2pcm
import baidu_ai
import play_mp3
#import FAQ
import opendb
while 1:
pyrec.rec("1.wav")
pcm_file = wav2pcm.wav_to_pcm("1.wav")
print(pcm_file)
res_str = baidu_ai.audio_to_text(pcm_file)
print(res_str)
#Q_str = FAQ.faq(res_str)
Q_str = opendb.poetmatch(res_str)
if Q_str != 0:
synth_file = baidu_ai.text_to_audio(Q_str)
#synth_file = baidu_ai.text_to_audio(res_str)
play_mp3.playmp3(synth_file)
#!/usr/bin/python3
import pyrec # 录音函数文件
import wav2pcm # wav转换pcm 函数文件
import baidu_ai # 语音合成函数,语音识别函数 文件
import playmp3 # 播放mp3 函数 文件
import FAQ
import tuling
import sizhi
pyrec.rec("1.wav") # 录音并生成wav文件,使用方式传入文件名
pcm_file = wav2pcm.wav_to_pcm("1.wav") # 将wav文件 转换成pcm文件 返回 pcm的文件名
res_str = baidu_ai.audio_to_text(pcm_file) # 将转换后的pcm音频文件识别成 文字 res_str
ter = sizhi.speak(res_str)
Q_str = str(ter)
synth_file = baidu_ai.text_to_audio(Q_str)
playmp3.play_mp3(synth_file)
import luyin
import wav2pcm
import shibie
import control
import time
import os
i=0
while True:
# 请说出语音指令
print("\n\n==================================================")
rec_yuyin = "/home/hsj/Documents/test"+str(i)+".wav"
#print(rec_yuyin)
luyin.audio_record(rec_yuyin, 3)
pcm_file = wav2pcm.wav_to_pcm(rec_yuyin)
asr_result = shibie.shibie(pcm_file) # 识别语音指令
if len(asr_result) != 0:
print("Identify Result:", asr_result)
print("Start Control...")
control.keyboard_control(asr_result) # 根据识别结果控制键盘
print("Control End...")
if asr_result.find("退出") != -1: # 如果是"退出"指令则结束程序
break;
time.sleep(1) # 延时1秒
i=i+1
from aip import AipSpeech
import time
import wav2pcm
#import io
#import sys
#sys.stdout=io.TextIOWrapper(sys.stdout.buffer,encoding='utf8')
APP_ID = '18240192'
API_KEY = 'l6olC3AktVQfKUksIh1Z8Dv3'
SECRET_KEY = '76sHQR5cUo9zigr1g7Ka4zkatNaGrNGT'
client = AipSpeech(APP_ID, API_KEY, SECRET_KEY)
pcm_file = wav2pcm.wav_to_pcm("D4_750.wav")
def get_file_content(filePath):
with open(filePath, 'rb') as fp:
return fp.read()
#识别本地文件
start_time = time.time()
ret = client.asr(get_file_content('F:\\p project\\3.7\\D4_750.wav'), 'pcm',
16000, {
'dev_pid': 1537,
})
used_time = time.time() - start_time
print("used time: {}s".format(round(time.time() - start_time, 2)))
print('ret:{}'.format(ret))