app文件
from flask import Flask, request, render_template, jsonify, send_file
from uuid import uuid4
import os
import asr_test app = Flask(__name__)
app.debug = True @app.route('/')
def index():
return render_template('index.html') @app.route('/uploader', methods=['POST'])
def uploader():
file = request.files.get('reco')
file_name = os.path.join('audio', f'{uuid4()}.wav')
file.save(file_name)
ret_filename = asr_test.my_ai(file_name)
print(ret_filename)
return jsonify({'filename': ret_filename}) @app.route('/get_audio/<filename>')
def get_audio(filename):
file = os.path.join('audio', filename)
return send_file(file) if __name__ == '__main__':
app.run('0.0.0.0', 5000)
调用百度语音识别与语音合成接口，把传来的语言识别成文字，并调用下面的相似度接口，返回回答的文字，然后利用语音合成返回回答
from aip import AipSpeech
import os
from my_npl import get_score
from uuid import uuid4 """ 你的 APPID AK SK """
APP_ID = '******'
API_KEY = '******'
SECRET_KEY = '******' client = AipSpeech(APP_ID, API_KEY, SECRET_KEY) # 读取文件
def get_file_content(filePath):
any2pcm_str = f"ffmpeg -y -i {filePath} -acodec pcm_s16le -f s16le -ac 1 -ar 16000 {filePath}.pcm"
os.system(any2pcm_str)
with open(f"{filePath}.pcm", 'rb') as fp:
return fp.read() # 识别本地文件
def my_ai(file):
res = client.asr(get_file_content(file), 'pcm', 16000, {
'dev_pid': 1536,
}) print(res.get('result'))
print(res)
question = res.get('result')[0]
req = get_score(question) req = client.synthesis(req, 'zh', 1, {
'vol': 5,
'pit': 5,
'spd': 4,
"per": 4
}) # 识别正确返回语音二进制 错误则返回dict 参照下面错误码
if not isinstance(req, dict):
ret_filename = f'{uuid4()}.mp3'
new_filename = os.path.join("audio", ret_filename)
with open(new_filename, 'wb') as f:
f.write(req)
return ret_filename
调用百度ai自然语言中的短文本相似度接口，使相似的问题得到相同的答案
from aip import AipNlp
from mytuling import to_tuling """ 你的 APPID AK SK """
APP_ID = '***'
API_KEY = '***'
SECRET_KEY = '***' client = AipNlp(APP_ID, API_KEY, SECRET_KEY) def get_score(Q):
if client.simnet(Q, '你叫什么名字').get('score') > 0.7:
return '我是大名鼎鼎的小王吧'
elif client.simnet(Q, '你今年几岁呀').get('score') > 0.7:
return '我今年已经1112岁啦'
else:
return to_tuling(Q)
调用图灵接口完成未设定问答的
import requests
tuling_url = 'http://openapi.tuling123.com/openapi/api/v2'
data = {"reqType": 0,
"perception": {
"inputText": {
"text": ""
}
}
,
"userInfo": {
"apiKey": "***",
"userId": "***"
}
}
def to_tuling(Q):
data["perception"]["inputText"]['text'] = Q
a = requests.post(url=tuling_url, json=data)
res = a.json()
print(res)
return res.get("results")[0].get("values").get("text")
简单前端页面
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<title>Title</title>
</head>
<body>
<audio src="" autoplay controls id="player"></audio>
<p>
<button onclick="start_reco()">开始录音</button>
</p>
<p>
<button onclick="stop_reco()">停止录音</button>
</p> </body>
<script src="/static/Recorder.js"></script>
<script src="/static/jQuery3.1.1.js"></script>
<script type="text/javascript">
var serv = "http://127.0.0.1:5000";
var audio_serv = serv + "/get_audio/";
var audio_context = new AudioContext();
navigator.getUserMedia = (navigator.getUserMedia ||
navigator.webkitGetUserMedia ||
navigator.mozGetUserMedia ||
navigator.msGetUserMedia); navigator.getUserMedia({audio: true}, create_stream, function (err) {
console.log(err)
}); function create_stream(user_media) {
var stream_input = audio_context.createMediaStreamSource(user_media);
reco = new Recorder(stream_input);
} function start_reco() {
reco.record();
} function stop_reco() {
reco.stop();
get_audio();
reco.clear();
} function get_audio() {
reco.exportWAV(function (wav_file) {
// wav_file 音频文件 Blob("wav","context")
console.log(wav_file);
var formdata = new FormData();
formdata.append("reco", wav_file);
$.ajax({
url: serv + "/uploader",
type: 'post',
processData: false,
contentType: false,
data: formdata,
dataType: 'json',
success: function (data) {
document.getElementById("player").src = audio_serv + data.filename;
}
}
);
})
}
</script>
</html>
前端页面
