Nodejs - IBM Watson Speech To Text Websocket connection error - node.js

Question

I have been trying to follow this official guide from IBM (https://developer.ibm.com/tutorials/add-a-trigger-word-to-your-watson-assistant/) to build a Nodejs voice assistant that answer after it recognizes a "wake up word". That guide seems a little bit outdated so I decided to use it with Assistant V2 and ibm-watson 5.2.0 package from npm.
I am getting a WebSocket connection error with not much information on it, the issue seems to be on line 33 and with the params I am sending to the 'recognizeUsingWebsocket' method. Am I missing something along this parameters?
const AssistantV2 = require('ibm-watson/assistant/v2');
const TextToSpeechV1 = require('ibm-watson/text-to-speech/v1');
const SpeechToTextV1 = require('ibm-watson/speech-to-text/v1');
const { IamAuthenticator } = require('ibm-watson/auth');
const mic = require('mic');
const conversation = new AssistantV2({
authenticator: new IamAuthenticator({ apikey: '<api_key>' }),
url: 'https://gateway-lon.watsonplatform.net/assistant/api/',
version: '2018-09-19'
});
const speechToText = new SpeechToTextV1({
authenticator: new IamAuthenticator({ apikey: '<api_key>' }),
serviceUrl: 'https://gateway-lon.watsonplatform.net/speech-to-text/api'
});
const textToSpeech = new TextToSpeechV1({
authenticator: new IamAuthenticator({ apikey: '<api_key>' })
});
const micParams = {
rate: 44100,
channels: 2,
debug: true,
exitOnSilence: 6
};
const microphone = mic(micParams);
const micInputStream = microphone.getAudioStream();
const textStream = micInputStream
.pipe(
speechToText.recognizeUsingWebSocket({
accessToken:'<access_token>',
contentType: 'audio/l16; rate=44100; channels=2',
interimResults: true,
inactivityTimeout: -1
})
)
.setEncoding('utf8');
const speakResponse = (text) => {
var params = {
text: text,
accept: 'audio/wav',
voice: 'en-US_AllisonVoice'
};
var writeStream = fs.createWriteStream('output.wav');
textToSpeech
.synthesize(params)
.then((audio) => {
audio.pipe(writeStream);
})
.catch((err) => {
console.log('error:', err);
});
writeStream.on('finish', function() {
ffprobe('output.wav', function(err, probeData) {
if (probeData) {
pauseDuration = probeData.format.duration;
microphone.pause();
speaker.play('output.wav');
startTime = new Date();
}
});
});
writeStream.on('error', function(err) {
console.log('Text-to-speech streaming error: ' + err);
});
};
function printContext(header) {
if (debug) {
console.log(header);
if (context.system) {
if (context.system.dialog_stack) {
const util = require('util');
console.log(" dialog_stack: ['" + util.inspect(context.system.dialog_stack, false, null) + "']");
}
}
}
}
function watsonSays(response) {
if (typeof response !== 'undefined') {
console.log('Watson says:', response);
}
}
function isActive(text) {
var elapsedTime = new Date() - startTime;
if (elapsedTime > SLEEP_TIME) {
// go to sleep
startTime = new Date();
botIsActive = false;
}
if (botIsActive) {
// in active conversation, so stay awake
startTime = new Date();
return true;
} else {
// we are asleep - did we get a wake up call?
if (text.toLowerCase().indexOf(wakeWord) > -1) {
// time to wake up
console.log('App just woke up');
botIsActive = true;
} else {
// false alarm, go back to sleep
console.log('App needs the wake up command');
}
return botIsActive;
}
}
function performConversation() {
console.log('App is listening, you may speak now.');
textStream.on('data', (user_speech_text) => {
userSpeechText = user_speech_text.toLowerCase();
console.log('\n\nApp hears: ', user_speech_text);
if (isActive(user_speech_text)) {
conversation.message(
{
assistantId: process.env.ASSISTANT_ID,
sessionId: process.env.SESSION_ID,
input: { text: user_speech_text }
},
(err, response) => {
console.log(err);
context = response.context;
watson_response = response.output.text[0];
if (watson_response) {
speakResponse(watson_response);
}
watsonSays(watson_response);
}
);
}
});
}
microphone.start();
performConversation();

Answer 1

It seems that you do not use the right endpoint for websockets :
note : api endpoints have had a new version in dec 2019
you use :
https://gateway-lon.watsonplatform.net/speech-to-text/api
and it should be something like : (i think the prefix wss is key)
wss://api.{location}.speech-to-text.watson.cloud.ibm.com/instances/{instance_id}/v1/recognize
cf Api reference : https://cloud.ibm.com/apidocs/speech-to-text/speech-to-text#websocket_methods