百度AI开放平台,语音识别,语音合成以及短文本相似度

百度AI开放平台:https://ai.baidu.com/

语音合成

from aip import AipSpeech
APP_ID="15420964"   #'你的 App ID'
API_KEY="6bPrLnkguN5ltxvfxRYP96Hk"  #'你的 Api Key'
SECRET_KEY="ckSFGccmaGr0b2EPGE3dueb1PkfW5IsW"   #'你的 Secret Key'

client=AipSpeech(APP_ID,API_KEY,SECRET_KEY)

result=client.synthesis("快乐的池塘里面有只小青蛙,呱呱呱儿  ","zh",1,{
    "vol":5,   #音量
    "spd":4,   #语速
    "pit":7,   #语调
    "per":1    #音色(0,1,3,4)
})
print(result)

if not isinstance(result,dict):
    with open("audio.mp3","wb") as f:
        f.write(result)

语音识别

from aip import AipSpeech
import os


""" 你的 APPID AK SK """
APP_ID="15420964"
API_KEY="6bPrLnkguN5ltxvfxRYP96Hk"
SECRET_KEY="ckSFGccmaGr0b2EPGE3dueb1PkfW5IsW"

client=AipSpeech(APP_ID,API_KEY,SECRET_KEY)


# 读取文件
def get_file_content(filePath):
    with open(filePath, 'rb') as fp:
        return fp.read()

# 识别本地文件
res=client.asr(get_file_content('wb.m4a'), 'pcm', 16000, {
    'dev_pid': 1536,
})

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

学说话

from aip import AipSpeech
import os
import time


""" 你的 APPID AK SK """

APP_ID="15420964"
API_KEY="6bPrLnkguN5ltxvfxRYP96Hk"
SECRET_KEY="ckSFGccmaGr0b2EPGE3dueb1PkfW5IsW"

client=AipSpeech(APP_ID,API_KEY,SECRET_KEY)

############语音识别##########
#读取文件
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()

# 识别本地文件
def audio2text(filename):
    res=client.asr(get_file_content(filename), 'pcm', 16000, {
        'dev_pid': 1536,
    })
    return res.get("result")[0]

############语音合成#########
def text2audio(getedtext):
    filename = f"{time.time()}.mp3"
    result = client.synthesis(getedtext, "zh", 1, {
        "vol": 5,
        "spd": 4,
        "pit": 7,
        "per": 0
    })

    if not isinstance(result, dict):
        with open(filename, "wb") as f:
            f.write(result)
    return   filename

getedtext=audio2text("wb.m4a")

res=text2audio(getedtext)

os.system(res)

语音回答问题

from aip import AipSpeech
import os
import time


""" 你的 APPID AK SK """

APP_ID="15420964"
API_KEY="6bPrLnkguN5ltxvfxRYP96Hk"
SECRET_KEY="ckSFGccmaGr0b2EPGE3dueb1PkfW5IsW"

client=AipSpeech(APP_ID,API_KEY,SECRET_KEY)

############语音识别##########
#读取文件
def get_file_content(filePath):
    with open(filePath, 'rb') as fp:
        return fp.read()

# 识别本地文件
def audio2text(filepath):
    res = client.asr(get_file_content(filepath), 'pcm', 16000, {
        'dev_pid': 1536,
    })
    return res.get("result")[0]

filename = f"{time.time()}.mp3"
############语音合成#########
def text2audio(getedtext):

    result = client.synthesis(getedtext, "zh", 1, {
        "vol": 5,
        "spd": 4,
        "pit": 7,
        "per": 1
    })
    # print(result)

    if not isinstance(result, dict):
        with open(filename, "wb") as f:
            f.write(result)
getedtext=audio2text("wb.m4a")

text2audio(getedtext)

os.system(filename)

短文本相似度

from aip import AipSpeech,AipNlp
import os
import time


""" 你的 APPID AK SK """

APP_ID="15420964"
API_KEY="6bPrLnkguN5ltxvfxRYP96Hk"
SECRET_KEY="ckSFGccmaGr0b2EPGE3dueb1PkfW5IsW"

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

############语音识别##########
#读取文件
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()

# 识别本地文件
def audio2text(filename):
    res=client.asr(get_file_content(filename), 'pcm', 16000, {
        'dev_pid': 1536,
    })
    return res.get("result")[0]

############语音合成#########
def text2audio(getedtext):
    filename = f"{time.time()}.mp3"
    result = client.synthesis(getedtext, "zh", 1, {
        "vol": 5,
        "spd": 4,
        "pit": 7,
        "per": 0
    })

    if not isinstance(result, dict):
        with open(filename, "wb") as f:
            f.write(result)
    return   filename

##########图灵################
def to_tuling(text):
    import requests

    args = {
        "reqType": 0,
        "perception": {
            "inputText": {
                "text": text
            }
        },
        "userInfo": {
            "apiKey": "e963f65c4c7a466a80e5aaa3510da2fa",
            "userId": "1111"
        }
    }

    url = "http://openapi.tuling123.com/openapi/api/v2"

    res = requests.post(url, json=args)
    print(res)
    text = res.json().get("results")[0].get("values").get("text")

    print("图灵答案",text)
    return text

###########调用################

getedtext=audio2text("wb.m4a")

if nlp.simnet("你叫什么名字",getedtext).get("score")>=0.68:
    getedtext="我才不告诉你呢,你个糟老头子坏得很"
else:
    getedtext=to_tuling(getedtext)

res=text2audio(getedtext)

os.system(res)

对话机器人玩具

应用结构:

在index.html中

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <title>Title</title>
</head>
<body>
<audio controls autoplay id="player"></audio>
<p>
    <button onclick="start_reco()" style="background-color: yellow">录制语音指令</button>
</p>
<p>
    <button onclick="stop_reco_audio()" style="background-color: blue">发送语音指令</button>
</p>
</body>
<!--<script type="application/javascript" src="/static/Recorder.js"></script>-->
<script type="application/javascript" src="https://cdn.bootcss.com/recorderjs/0.1.0/recorder.js"></script>
<script type="text/javascript" src="/static/jQuery3.1.1.js"></script>

<script type="text/javascript">
    var reco = null;
    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_audio() {
        reco.stop();
        send_audio();
        reco.clear();
    }


    function send_audio() {
        reco.exportWAV(function (wav_file) {
            var formdata = new FormData();
            formdata.append("record", wav_file);
            console.log(formdata);
            $.ajax({
                url: "http://192.168.43.158:9527/ai",
                type: 'post',
                processData: false,
                contentType: false,
                data: formdata,
                dataType: 'json',
                success: function (data) {
                    document.getElementById("player").src ="http://192.168.43.158:9527/get_audio/" + data.filename
                }
            });

        })
    }



</script>
</html>

在app.py中

from flask import Flask,render_template,request,jsonify,send_file
from uuid import uuid4
import baidu_ai

app = Flask(__name__)

@app.route("/")
def index():
    return render_template("index.html")

@app.route("/ai",methods=["POST"])
def ai():
    # 1.保存录音文件
    print(111)
    audio = request.files.get("record")
    print('audio',audio)
    filename = f"{uuid4()}.wav"
    audio.save(filename)
    #2.将录音文件转换为PCM发送给百度进行语音识别
    q_text = baidu_ai.audio2text(filename)
    print(8585)
    print(q_text)
    #3.将识别的问题交给图灵或自主处理获取答案
    a_text = baidu_ai.to_tuling(q_text)
    print(a_text)
    #4.将答案发送给百度语音合成，合成音频文件
    a_file = baidu_ai.text2audio(a_text)
    print(a_file)
    #5.将音频文件发送给前端播放

    return jsonify({"filename":a_file})


@app.route("/get_audio/<filename>")
def get_audio(filename):
    print(filename)
    return send_file(filename)



if __name__ == '__main__':
    app.run("0.0.0.0",9527,debug=True)

在baidu_ai.py中

from aip import AipSpeech,AipNlp
import time,os

""" 你的 APPID AK SK """
APP_ID="15420964"
API_KEY="6bPrLnkguN5ltxvfxRYP96Hk"
SECRET_KEY="ckSFGccmaGr0b2EPGE3dueb1PkfW5IsW"

# client = AipSpeech(APP_ID, API_KEY, SECRET_KEY)

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

# 读取文件
def get_file_content(filePath):
    print(234)
    os.system(f"ffmpeg -y  -i {filePath} -acodec pcm_s16le -f s16le -ac 1 -ar 16000 {filePath}.pcm")
    print(354)
    with open(f"{filePath}.pcm", 'rb') as fp:
        return fp.read()

def audio2text(filepath):
    # 识别本地文件
    print(123)
    res = client.asr(get_file_content(filepath), 'pcm', 16000, {
        'dev_pid': 1536,
    })
    print(456456)
    print('----------res',res)

    return res.get("result")[0]

def text2audio(text):
    filename = f"{time.time()}.mp3"
    result = client.synthesis(text, 'zh', 1, {
        'vol': 5,
        "spd": 3,
        "pit": 7,
        "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(result, dict):
        with open(filename, 'wb') as f:
            f.write(result)

    return filename

def to_tuling(text):
    import requests

    args = {
        "reqType": 0,
        "perception": {
            "inputText": {
                "text": text
            }
        },
        "userInfo": {
            "apiKey": "e963f65c4c7a466a80e5aaa3510da2fa",
            "userId": "1111"
        }
    }

    url = "http://openapi.tuling123.com/openapi/api/v2"

    res = requests.post(url, json=args)
    print(res)
    text = res.json().get("results")[0].get("values").get("text")

    print("图灵答案",text)
    return text