人工智能---语音合成\识别\图灵机械人语音交互

 1 """ 你的 APPID AK SK """
 2 import os
 3 from aip import AipSpeech, AipNlp
 4 from ss import tuling
 5 from ss.录音 import rec
 6 
 7 APP_ID = ‘16027699‘
 8 API_KEY = ‘rZrseMjmXXHadrp0fbUAyb6Z‘
 9 SECRET_KEY = ‘0tFVGEW09PpPCMxXBqVrda59fQf3wY3c‘
10 
11 client = AipSpeech(APP_ID, API_KEY, SECRET_KEY)
12 client_NLp = AipNlp(APP_ID, API_KEY, SECRET_KEY)
13 
14 
15 #文字合成语音
16 def text_audio(text):
17     result = client.synthesis(text, ‘zh‘, 1, {
18         ‘vol‘: 5,
19         ‘per‘: 4,
20         ‘pit‘: 8,
21         ‘spd‘: 5,
22     })
23 
24     # 识别正确返回语音二进制 错误则返回dict 参照下面错误码
25     if not isinstance(result, dict):
26         with open(‘audios1.mp3‘, ‘wb‘) as f:
27             f.write(result)
28     return ‘audios1.mp3‘
29 
30 #语音识别--文字
31 def audio_text(filename):
32     # 识别本地文件
33     res = client.asr(get_file_content(filename), ‘pcm‘, 16000, {
34         ‘dev_pid‘: 1536,
35     })
36     return res.get(‘result‘)[0]
37 
38 
39 # -转换格式--读取文件
40 def get_file_content(filePath):
41     os.system(f"ffmpeg -y  -i {filePath}.wav  -acodec pcm_s16le -f s16le -ac 1 -ar 16000 {filePath}.pcm ")
42     with open(f‘{filePath}.pcm‘, ‘rb‘) as fp:
43         return fp.read()
44 
45 
46 if __name__ == ‘__main__‘:
47 
48     re=rec(‘file_name.wav‘)  #录音,形成录音文件
49     text=audio_text(‘file_name‘)  #录音转为文字
50     # print(text)
51     # score =client_NLp.simnet(text,‘今天天气‘).get(‘score‘)
52     # print(score)
53     # if score>0.58:
54     #     print(‘1111‘)
55     #     filename=text_audio(‘是呀,撩起呀!‘)
56     #     print(filename)
57     #     # os.system(f"ffplay {filename}")
58     #     os.system(filename)
59 
60     answer=tuling.get_help_from_tuling(text,‘mabing‘) #文字信息传递给tuling得到答复信息
61     filename = text_audio(answer)  #答复信息再转为语音
62     os.system(filename)            #播放

 1 import requests
 2 
 3 def get_help_from_tuling(text,uid):
 4     URL = ‘http://openapi.tuling123.com/openapi/api/v2‘
 5 
 6     data = {
 7         "perception": {
 8             "inputText": {
 9                 "text": ‘‘
10             }
11         },
12         "userInfo": {
13             "apiKey": "86fd93bd29e941eebe314766616a3f15",
14             "userId": "123456"
15         }
16     }
17     data["perception"]["inputText"]["text"] = text
18     data["userInfo"]["userId"] = uid
19     res = requests.post(URL, json=data)
20     print(res.json())
21     return res.json().get(‘results‘)[0].get(‘values‘).get(‘text‘)

 1 import pyaudio
 2 import wave
 3 
 4 CHUNK = 1024
 5 FORMAT = pyaudio.paInt16
 6 CHANNELS = 2
 7 RATE = 16000
 8 RECORD_SECONDS = 6
 9 # WAVE_OUTPUT_FILENAME = "Oldboy.wav"
10 def rec(file_name):
11     p = pyaudio.PyAudio()
12 
13     stream = p.open(format=FORMAT,
14                     channels=CHANNELS,
15                     rate=RATE,
16                     input=True,
17                     frames_per_buffer=CHUNK)
18 
19     print("开始录音,请说话......")
20 
21     frames = []
22 
23     for i in range(0, int(RATE / CHUNK * RECORD_SECONDS)):
24         data = stream.read(CHUNK)
25         frames.append(data)
26 
27     print("录音结束,请闭嘴!")
28 
29     stream.stop_stream()
30     stream.close()
31     p.terminate()
32 
33     wf = wave.open(file_name, ‘wb‘)
34     wf.setnchannels(CHANNELS)
35     wf.setsampwidth(p.get_sample_size(FORMAT))
36     wf.setframerate(RATE)
37     wf.writeframes(b‘‘.join(frames))
38     wf.close()