我的用例:
我的例子是,我正在制作一个用于收听播客的机器人,在这个播客中,用户将调用twilio号,bot将询问您想听哪种播客,然后记录10秒
当录制完时,它说当我们找到播客时,用户请稍候。
我想在我的web钩子中记录,这样我就可以找出来电者的心情,并从我的数据库中找到合适的播客mp3文件,然后播放给呼叫者。
我面临的问题:
我所有的钩子都是空的
我的代码:
var express = require("express");
var bodyParser = require("body-parser");
var VoiceResponse = require('twilio').twiml.VoiceResponse;
var app = express();
var port = (process.env.PORT || 4000);
app.use(bodyParser.json())
// helper to append a new "Say" verb with alice voice
function say(text, twimlRef) {
twimlRef.say({ voice: 'alice' }, text);
}
// respond with the current TwiML content
function respond(responseRef, twimlRef) {
responseRef.type('text/xml');
responseRef.send(twimlRef.toString());
}
app.post("/voice", function (request, response, next) {
console.log("request: ", request.body); //body is comming as empty object
var phone = request.body.From;
var input = request.body.RecordingUrl;
var twiml = new VoiceResponse();
console.log("phone, input: ", phone, input);
say('What type of podcast would you like to listen. Press any key to finish.', twiml);
twiml.record({
method: 'POST',
action: '/voice/transcribe',
transcribeCallback: '/voice/transcribe',
maxLength: 10
});
respond(response, twiml);
});
app.post("/voice/transcribe", function (request, response, next) {
console.log("request: ", request.body); //body is comming as empty object
var phone = request.body.From;
var input = request.body.RecordingUrl;
var twiml = new VoiceResponse();
var transcript = request.body.TranscriptionText;
console.log("transcribe text: ", transcript);
//here i will do some magic(Ai) to detect user mood and find an
//appropriate mp3 file from my database and send to twilio
var mp3Url = 'https://api.twilio.com/cowbell.mp3'
say('start playing.', twiml);
twiml.play(mp3Url);
respond(response, twiml);
});
app.listen(port, function () {
console.log('app is running on port', port);
});邮递员的API测试:


在twilio:上添加了url作为web钩子

Heroku日志:

发布于 2017-05-19 16:25:15
两位开发人员在这里传道。
您正在使用body-parser,这是很好的。但是,您正在使用JSON解析器。Twilio以application/www-x-form-urlencoded格式进行请求,因此您应该更改:
app.use(bodyParser.json())至
app.use(bodyParser.urlencoded({ extended: false }))然后,您应该将解析的主体视为request.body对象的一部分。
另外,transcribeCallback被异步地发送到调用。因此,响应该请求返回TwiML根本不会影响调用。当你得到转录结果时,你需要修改飞行中的呼叫,将其重定向到某个新的TwiML。使用Node.js更新调用如下所示的一个例子
const accountSid = 'your_account_sid';
const authToken = 'your_auth_token';
const client = require('twilio')(accountSid, authToken);
client.calls('CAe1644a7eed5088b159577c5802d8be38')
.update({
url: 'http://demo.twilio.com/docs/voice.xml',
method: 'POST',
})
.then((call) => console.log(call.to));https://stackoverflow.com/questions/43861788
复制相似问题