码迷,mamicode.com
首页 > 其他好文 > 详细

人工智能 - 基于火狐浏览器的语音识别,语音自动回复

时间:2019-04-07 21:52:09      阅读:184      评论:0      收藏:0      [点我收藏+]

标签:insert   content   web   system   index   代码   rip   r.js   main   

人工智能 - 基于火狐浏览器的语音识别,语音自动回复

一. 需求工具

  1. 下载安装火狐浏览器

    因为火狐浏览器自带麦克风

  2. 安装Recorer.js

    一个易于使用的录音机,以Matt Diamond的 录音机为核心

    http://www.chengfeilong.com/recorderjs-info

  3. 百度AI

  4. 图灵机器人

  5. mongoDB(数据库) - pymongo

    可以选择其他数据库

  6. jQuery

  7. Flask框架

二. 不废话,上代码

1. index.HTML

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <title>我是玩具</title>
</head>
<body>

<p><audio id="player" controls autoplay ></audio></p>
<button onclick="start_reco()">录音</button>
<button onclick="stop_reco()">发送语音</button>
<div id="content"></div>


</body>
<script type="text/javascript" src="/static/Recorder.js"></script>
<script type="text/javascript" src="/static/jquery-3.3.1.min.js"></script>
<script type="text/javascript">
    var serv = "http://192.168.11.206:9527";

    var reco = null;
    var audio_context = new AudioContext();//音频内容对象
    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();

        reco.exportWAV(function (wav_file) {
            console.log(wav_file);
            var formdata = new FormData(); // form 表单 {key:value}
            formdata.append("reco",wav_file); // form input type="file"
            formdata.append("key","value");
            // # <input type="text" name = "key"> value
            $.ajax({
            url: serv + "/upload",
            type: 'post',
            processData: false,
            contentType: false,
            data: formdata,
            dataType: 'json',
            success: function (data) {
                console.log(data);
                if(data.code == 0){
                    document.getElementById("player").src = "http://192.168.11.206:9527/get_file/"+data.filename;
                    document.getElementById("content").innerText = data.content;

                }
            }
            })
        });

        reco.clear();
    }

</script>
</html>

2. adiou.py(封装的 百度AI, 图灵机器人 函数)

from aip import AipSpeech
from aip import AipNlp
import os
from uuid import uuid4

""" 你的 APPID AK SK """
APP_ID = '15837844'
API_KEY = '411VNGbuZVbDNZU78LqTzfsV'
SECRET_KEY = '84AnwR2NARGMqnC6WFnzqQL9WWdWh5bW'

client = AipSpeech(APP_ID, API_KEY, SECRET_KEY)
nlp_client = AipNlp(APP_ID, API_KEY, SECRET_KEY)


def get_file_content(filePath):
    os.system(f"ffmpeg -y  -i {filePath}  -acodec pcm_s16le -f s16le -ac 1 -ar 16000 {filePath}.pcm")
    with open(f"{filePath}.pcm", 'rb') as fp:
        return fp.read()


def audio2text(filePath):
    res = client.asr(get_file_content(filePath), 'pcm', 16000, {
        'dev_pid': 1536,
    })

    text = res.get("result")[0]
    print(text)

    return text

import requests


def to_tuling(text, uid):
    data = {
        "perception": {
            "inputText": {
                "text": "北京"
            }
        },
        "userInfo": {
            "apiKey": "a4c4a668c9f94d0c928544f95a3c44fb",
            "userId": "123"
        }
    }
    data["perception"]["inputText"]["text"] = text
    data["userInfo"]["userId"] = uid
    res = requests.post("http://openapi.tuling123.com/openapi/api/v2", json=data)
    # print(res.content)
    res_json = res.json()

    text = res_json.get("results")[0].get("values").get("text")
    print(text)
    return text

def my_nlp(text):
    if nlp_client.simnet(text,"你叫什么名字").get("score") >= 0.75:
        A = "我叫银王八"
        return A

    if nlp_client.simnet(text,"你今年几岁了").get("score") >= 0.75:
        A = "我今年999岁了"
        return A

    A = to_tuling(text,"open123")
    return A


def text2audio(text):
    result = client.synthesis(text, 'zh', 1, {
        'vol': 5,
        'per': 4,
        'spd': 4,
        'pit': 7,
    })

    filename = f"{uuid4()}.mp3"

    # 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(result, dict):
        # print(result)
        with open(filename, 'wb') as f:
            f.write(result)


    return filename

3. app.py(路由视图, 逻辑)****

    from flask import Flask, render_template, request, jsonify, send_file
    from uuid import uuid4
    from adiou import audio2text, text2audio, my_nlp
    from mongodb import MONGODB
    from flask_cors import CORS
    
    
    app = Flask(__name__)
    CORS(app,  resources={r"/*": {"origins": "*"}})
    CORS(app, supports_credentials=True)
    
    
    @app.route('/')
    def hello_world():
        print('////')
        return render_template("index.html")
    
    
    @app.route("/upload", methods=["POST"])
    def upload():
        print("111")
        fi = request.files.get("reco")
        print("2222", fi)
        fi_name = f"{uuid4()}.wav"
        print(fi_name, "3333")
        fi.save(fi_name)
    
        text = audio2text(fi_name)
        print(text, "text")
        text1 = {"kong": text}
        res1 = MONGODB.users.insert_one(text1)
        print(res1)
    
        new_text = my_nlp(text)
        print(new_text, "new_text")
        text2 = {"机器人": new_text}
        res2 = MONGODB.users.insert_one(text2)
        print(res2)
    
        filename = text2audio(new_text)
        print(filename, "filename")
    
    
    
    
        ret = {
            "filename":filename,
            "content":new_text,
            "code":0
        }
    
        return jsonify(ret)
    
    
    @app.route("/get_file/<filename>")
    def get_file(filename):
        return send_file(filename)

4. mongodb.py

from pymongo import MongoClient

conn = MongoClient("127.0.0.1", 27017)
MONGODB = conn["db3"]

5. run.py(Flask启动)

from app import app


if __name__ == '__main__':

    app.run("0.0.0.0", 9527, debug=True)

人工智能 - 基于火狐浏览器的语音识别,语音自动回复

标签:insert   content   web   system   index   代码   rip   r.js   main   

原文地址:https://www.cnblogs.com/konghui/p/10667235.html

(0)
(0)
   
举报
评论 一句话评论(0
登录后才能评论!
© 2014 mamicode.com 版权所有  联系我们:gaon5@hotmail.com
迷上了代码!