WS & TTS urls configurable

.devcontainer/devcontainer.json

@@ -16,7 +16,11 @@
 	 //"forwardPorts": [ "28080:8080", "28081:8081"],
 	// tell vscode to forward the port 8080 to the container to random local port
 	"appPort": ["28080:8080", "28081:8081"],
-
+	"remoteEnv": {
+		"TTS_BACKEND_URL": "http://192.168.0.10:9008/asr",
+		//"WS_URL":"ws://192.168.0.10:28081"
+		"WS_URL":"wss://ws.ai.d-popov.com"
+	  }
 	// Use 'postCreateCommand' to run commands after the container is created.
 	// "postCreateCommand": "npm install ws express request",
 

Dockerfile

@@ -10,8 +10,8 @@ RUN echo $USERNAME ALL=\(root\) NOPASSWD:ALL > /etc/sudoers.d/$USERNAME \
     && chmod 0440 /etc/sudoers.d/$USERNAME
 
 # Set `DEVCONTAINER` environment variable to help with orientation
-ENV DEVCONTAINER=true
+#ENV DEVCONTAINER=true
-
+#! env declarations not copied to devcontainer
 
 ########## Modified Dockerfile ##########
 

package-lock.json

@@ -8,9 +8,11 @@
             "name": "kevin-ai",
             "version": "1.0.0",
             "dependencies": {
+                "dotenv": "^16.0.3",
                 "express": "^4.18.2",
                 "request": "^2.88.2",
-                "ws": "^8.12.1"
+                "ws": "^8.12.1",
+                "wscat": "^5.2.0"
             }
         },
         "node_modules/accepts": {
@@ -25,6 +27,38 @@
                 "node": ">= 0.6"
             }
         },
+        "node_modules/agent-base": {
+            "version": "6.0.2",
+            "resolved": "https://registry.npmjs.org/agent-base/-/agent-base-6.0.2.tgz",
+            "integrity": "sha512-RZNwNclF7+MS/8bDg70amg32dyeZGZxiDuQmZxKLAlQjr3jGyLx+4Kkk58UO7D2QdgFIQCovuSuZESne6RG6XQ==",
+            "dependencies": {
+                "debug": "4"
+            },
+            "engines": {
+                "node": ">= 6.0.0"
+            }
+        },
+        "node_modules/agent-base/node_modules/debug": {
+            "version": "4.3.4",
+            "resolved": "https://registry.npmjs.org/debug/-/debug-4.3.4.tgz",
+            "integrity": "sha512-PRWFHuSU3eDtQJPvnNY7Jcket1j0t5OuOsFzPPzsekD52Zl8qUfFIPEiswXqIvHWGVHOgX+7G/vCNNhehwxfkQ==",
+            "dependencies": {
+                "ms": "2.1.2"
+            },
+            "engines": {
+                "node": ">=6.0"
+            },
+            "peerDependenciesMeta": {
+                "supports-color": {
+                    "optional": true
+                }
+            }
+        },
+        "node_modules/agent-base/node_modules/ms": {
+            "version": "2.1.2",
+            "resolved": "https://registry.npmjs.org/ms/-/ms-2.1.2.tgz",
+            "integrity": "sha512-sGkPx+VjMtmA6MX27oA4FBFELFCZZ4S4XqeGOXCv68tT+jb3vk/RyaKWP0PTKyWtmLSM0b+adUTEvbs1PEaH2w=="
+        },
         "node_modules/ajv": {
             "version": "6.12.6",
             "resolved": "https://registry.npmjs.org/ajv/-/ajv-6.12.6.tgz",
@@ -146,6 +180,14 @@
                 "node": ">= 0.8"
             }
         },
+        "node_modules/commander": {
+            "version": "9.5.0",
+            "resolved": "https://registry.npmjs.org/commander/-/commander-9.5.0.tgz",
+            "integrity": "sha512-KRs7WVDKg86PWiuAqhDrAQnTXZKraVcCc6vFdL14qrZ/DcWwuRo7VoiYXalXO7S5GKpqYiVEwCbgFDfxNHKJBQ==",
+            "engines": {
+                "node": "^12.20.0 || >=14"
+            }
+        },
         "node_modules/content-disposition": {
             "version": "0.5.4",
             "resolved": "https://registry.npmjs.org/content-disposition/-/content-disposition-0.5.4.tgz",
@@ -227,6 +269,14 @@
                 "npm": "1.2.8000 || >= 1.4.16"
             }
         },
+        "node_modules/dotenv": {
+            "version": "16.0.3",
+            "resolved": "https://registry.npmjs.org/dotenv/-/dotenv-16.0.3.tgz",
+            "integrity": "sha512-7GO6HghkA5fYG9TYnNxi14/7K9f5occMlp3zXAuSxn7CKCxt9xbNWG7yF8hTCSUchlfWSe3uLmlPfigevRItzQ==",
+            "engines": {
+                "node": ">=12"
+            }
+        },
         "node_modules/ecc-jsbn": {
             "version": "0.1.2",
             "resolved": "https://registry.npmjs.org/ecc-jsbn/-/ecc-jsbn-0.1.2.tgz",
@@ -478,6 +528,39 @@
                 "npm": ">=1.3.7"
             }
         },
+        "node_modules/https-proxy-agent": {
+            "version": "5.0.1",
+            "resolved": "https://registry.npmjs.org/https-proxy-agent/-/https-proxy-agent-5.0.1.tgz",
+            "integrity": "sha512-dFcAjpTQFgoLMzC2VwU+C/CbS7uRL0lWmxDITmqm7C+7F0Odmj6s9l6alZc6AELXhrnggM2CeWSXHGOdX2YtwA==",
+            "dependencies": {
+                "agent-base": "6",
+                "debug": "4"
+            },
+            "engines": {
+                "node": ">= 6"
+            }
+        },
+        "node_modules/https-proxy-agent/node_modules/debug": {
+            "version": "4.3.4",
+            "resolved": "https://registry.npmjs.org/debug/-/debug-4.3.4.tgz",
+            "integrity": "sha512-PRWFHuSU3eDtQJPvnNY7Jcket1j0t5OuOsFzPPzsekD52Zl8qUfFIPEiswXqIvHWGVHOgX+7G/vCNNhehwxfkQ==",
+            "dependencies": {
+                "ms": "2.1.2"
+            },
+            "engines": {
+                "node": ">=6.0"
+            },
+            "peerDependenciesMeta": {
+                "supports-color": {
+                    "optional": true
+                }
+            }
+        },
+        "node_modules/https-proxy-agent/node_modules/ms": {
+            "version": "2.1.2",
+            "resolved": "https://registry.npmjs.org/ms/-/ms-2.1.2.tgz",
+            "integrity": "sha512-sGkPx+VjMtmA6MX27oA4FBFELFCZZ4S4XqeGOXCv68tT+jb3vk/RyaKWP0PTKyWtmLSM0b+adUTEvbs1PEaH2w=="
+        },
         "node_modules/iconv-lite": {
             "version": "0.4.24",
             "resolved": "https://registry.npmjs.org/iconv-lite/-/iconv-lite-0.4.24.tgz",
@@ -602,6 +685,11 @@
             "resolved": "https://registry.npmjs.org/ms/-/ms-2.0.0.tgz",
             "integrity": "sha512-Tpp60P6IUJDTuOq/5Z8cdskzJujfwqfOTkrwIwj7IRISpnkJnT6SyJ4PCPnGMoFjC9ddhal5KVIYtAt97ix05A=="
         },
+        "node_modules/mute-stream": {
+            "version": "0.0.8",
+            "resolved": "https://registry.npmjs.org/mute-stream/-/mute-stream-0.0.8.tgz",
+            "integrity": "sha512-nnbWWOkoWyUsTjKrhgD0dcz22mdkSnpYqbEjIm2nhwhuxlSkpywJmBo8h0ZqJdkp73mb90SssHkN4rsRaBAfAA=="
+        },
         "node_modules/negotiator": {
             "version": "0.6.3",
             "resolved": "https://registry.npmjs.org/negotiator/-/negotiator-0.6.3.tgz",
@@ -716,6 +804,17 @@
                 "node": ">= 0.8"
             }
         },
+        "node_modules/read": {
+            "version": "1.0.7",
+            "resolved": "https://registry.npmjs.org/read/-/read-1.0.7.tgz",
+            "integrity": "sha512-rSOKNYUmaxy0om1BNjMN4ezNT6VKK+2xF4GBhc81mkH7L60i6dp8qPYrkndNLT3QPphoII3maL9PVC9XmhHwVQ==",
+            "dependencies": {
+                "mute-stream": "~0.0.4"
+            },
+            "engines": {
+                "node": ">=0.8"
+            }
+        },
         "node_modules/request": {
             "version": "2.88.2",
             "resolved": "https://registry.npmjs.org/request/-/request-2.88.2.tgz",
@@ -992,6 +1091,20 @@
                     "optional": true
                 }
             }
+        },
+        "node_modules/wscat": {
+            "version": "5.2.0",
+            "resolved": "https://registry.npmjs.org/wscat/-/wscat-5.2.0.tgz",
+            "integrity": "sha512-UkVzuBdv3jk1Nt0mVCTw0wt/2kGPXry9MZMMUHYE/kEIJdtz1Ez28HD2WQdapC75tM10KZVL8EHG1/WHFK9dtw==",
+            "dependencies": {
+                "commander": "^9.3.0",
+                "https-proxy-agent": "^5.0.0",
+                "read": "^1.0.7",
+                "ws": "^8.0.0"
+            },
+            "bin": {
+                "wscat": "bin/wscat"
+            }
         }
     }
 }

web/client.html

@@ -1,7 +1,7 @@
 <!DOCTYPE html>
 <html>
-  <head>
+
-    <meta charset="UTF-8">
+<head>
     <title>Real-time Speech-to-Text</title>
     <style>
         .recording {
@@ -9,74 +9,260 @@
             color: white;
         }
     </style>
-  </head>
+</head>
-  <body>
+
-    <h1>Real-time Speech-to-Text</h1>
+<body>
-    <p id="connection-status">WebSocket disconnected</p>
+    <h1>Rt STT</h1>
-    <button id="record-button" disabled > Start Recording</button>
+    <label class="toggle">
+        <input type="checkbox"
+            id="autosend" />
+        <span class="slider">Continious</span>
+    </label>
+    <button id="record-button"
+        disabled>Start Recording</button>
+    <span id="connection-status"></span>
+
+    <div id="status-recording"></div>
     <p id="transcription"></p>
+
+    <canvas id="canvas"
+        width="500"
+        height="500"></canvas>
     <script>
-        let socket = new WebSocket('ws://localhost:8081');
+        let socket;
         let audioRecorder;
         let recording = false;
         let recordButton;
-        let connectionStatus;
+        let connected = false;
+        let connectionStatus; //HTML auto generated
+        let statusRecording; //HTML auto generated
+        let audioContext;
+        let serverTime;
 
-        function connect(socket) {
+        let volumeChecker;
-            console.log("trying to connect to server");
+        let lastVolumes = new Array(5);
+        let averageVolume;
+        let silenceCount = 0;
+        let isSpeaking = false;
+        let soundDetected = false;
+        let speakingCount = 0;
+        let SILENCE_DELAY_MS = 100;
+        const volumeHistory = [];
+
+        let canvas = document.getElementById("canvas");
+        let canvasCtx = canvas.getContext("2d");
+        let barWidth = 10;
+        let barSpacing = 5;
+
+        // Draw sliding bar graph
+        function drawSlidingBarGraph(lastVolumes) {
+            canvasCtx.clearRect(0, 0, canvas.width, canvas.height);
+            // Draw bars
+            for (let i = 0; i < lastVolumes.length; i++) {
+                let value = lastVolumes[i];
+                let barHeight = (value / 255) * canvas.height;
+                let x = i * (barWidth + barSpacing);
+                let y = canvas.height - barHeight;
+                canvasCtx.fillRect(x, y, barWidth, barHeight);
+            }
+        }
+
+        // Check the audio level every SILENCE_DELAY_MS milliseconds
+        volumeChecker = setInterval(() => {
+            if (!audioContext) {
+                console.log("No audio context");
+                return;
+            }
+            const frequencyData = new Uint8Array(analyser.frequencyBinCount);
+            //analyser.getByteTimeDomainData(dataArray);//history
+            analyser.getByteFrequencyData(frequencyData); //current
+
+            let totalVolume = 0;
+            for (let i = 0; i < frequencyData.length; i++) {
+                totalVolume += frequencyData[i];
+            }
+            averageVolume = totalVolume / frequencyData.length;
+
+            volumeHistory.push(averageVolume);
+            if (volumeHistory.length > 100) {
+                volumeHistory.shift();
+            }
+
+            const threshold = volumeHistory.reduce((acc, curr) => acc + curr) / volumeHistory.length + 5;
+            const isSilent = averageVolume < threshold;
+
+            // count speaking and silence
+            if (averageVolume > threshold) {
+                if (speakingCount == 0 && audioRecorder) {
+                    console.log("startint new recording");
+                    audioRecorder.stop();
+                    audioRecorder.start();
+                    soundDetected = true;
+                }
+                speakingCount++;
+                if (speakingCount > 7) {
+                    statusRecording.innerHTML = "Listening...";
+                    statusRecording.style.color = "green";
+                    isSpeaking = true;
+                    console.log("Was silent and is now speaking. (" + averageVolume + " averageVolume).");
+                }
+            } else {
+                speakingCount = 0;
+                if (isSpeaking) {
+                    silenceCount++;
+                    if (silenceCount > 5) {
+                        if (autosend.checked) {
+                            console.log("Was speakng and is now silence. (" + averageVolume + " averageVolume). Sending audio to server.");
+                            audioRecorder.stop();
+                            audioRecorder.start();
+                        }
+                        isSpeaking = false;
+                        statusRecording.innerHTML = "Silence detected...";
+                        statusRecording.style.color = "orange";
+                    }
+                }
+            }
+
+            //console.log(`Average volume: ${averageVolume}, isSilent: ${isSilent}, threshold: ${threshold}`);
+            //drawSlidingBarGraph(lastVolumes);
+        }, SILENCE_DELAY_MS);
+
+        function InitAudioAnalyser(stream) {
+            // Initialize the Web Audio API
+            audioContext = new AudioContext();
+            const source = audioContext.createMediaStreamSource(stream);
+            analyser = audioContext.createAnalyser();
+            //analyser.fftSize = 32;
+            analyser.fftSize = 2048;
+            analyser.smoothingTimeConstant = 0.8;
+            source.connect(analyser);
+            console.log("Audio context initialized. analyser sampling: " + audioContext.sampleRate + "Hz, recorder sampling: " + audioRecorder.sampleRate + "Hz");
+        }
+
+        function connect() {
+            connectionStatus.innerHTML = "Connecting to WS...";
+            let wsurl = "ws://localhost:8081";
+            //get crrent ws url from the server
+            fetch("/wsurl")
+                .then((response) => response.text())
+                .then((data) => {
+                    wsurl = data;
+                    console.log("Got ws url: '" + wsurl + "'");
+
+                })
+                .then(() => {
+                    console.log("connecting to '" + wsurl + "'...");
+                    socket = new WebSocket(wsurl);
                     socket.onopen = () => {
                         console.log("WebSocket connection opened.");
-                connectionStatus.innerHTML = "Connected";
+                        connectionStatus.innerHTML = "Connected to " + wsurl;
+                        transcription.innerHTML = "Whisper something into the microphone...";
                         recordButton.disabled = false;
+                        connected = true;
                     };
-            socket.onmessage = (event) => {
+                    socket.onmessage = onmessage;
-                transcription.innerHTML = event.data;
-            };
                     socket.onclose = () => {
                         console.log("WebSocket connection closed");
                         connectionStatus.innerHTML = "Disconnected";
                         recordButton.disabled = true;
-                setTimeout(connect, 5000);
+                        connected = false;
+                        setTimeout(() => {
+                            connect();
+                        }, 5000);
                     };
+                })
+                .catch((error) => {
+                    console.log("Error getting ws url: " + error);
+                    connectionStatus.innerHTML = "Error getting ws url: " + error;
+                });
+        };
+
+
+        function onmessage(event) {
+            let latency = Date.now() - serverTime;
+            console.log("Received message from server: " + event.data + " (latency: " + latency + "ms)");
+            if (autosend.checked) {
+                //append to the text on new line
+                transcription.innerHTML += "<br>>" + event.data;
+                statusRecording.innerHTML = "waiting...";
+                statusRecording.style.color = "black";
+            } else {
+                //replace the text
+                transcription.innerHTML = event.data;
+            }
         }
 
-        function startRecording() {
+        function startListening() {
+            // Initialize canvas
+            canvasCtx.fillStyle = "green";
             recording = true;
-            navigator.mediaDevices.getUserMedia({ audio: true })
+            navigator.mediaDevices.getUserMedia({ audio: { sampleRate: 16000, echoCancellation: true } }).then((stream) => {
-                .then((stream) => {
                 audioRecorder = new MediaRecorder(stream);
                 audioRecorder.start();
-                    audioRecorder.addEventListener('dataavailable', (event) => {
+                audioRecorder.addEventListener("dataavailable", (event) => {
-                        console.log("Sent some audio data to server.");
+                    console.log("Audio data available: " + event.data.size + " bytes");
-                        socket.send(event.data);
+                    if (!soundDetected && autosend.checked) {
+                        console.log("discarding audio data because not speaking");
+                        return;
+                    }
+                    if (event.data.size > 0) {
+                        sendAudioToServer(event.data);
+                        soundDetected = false;
+                    }
                 });
+
+                InitAudioAnalyser(stream);
             });
-            recordButton.classList.add("recording");
+
             recordButton.innerHTML = "Stop Recording";
+            recordButton.classList.add("recording");
         }
 
-        function stopRecording() {
+        function stopListening() {
             recording = false;
             audioRecorder.stop();
-            recordButton.classList.remove("recording");
             recordButton.innerHTML = "Start Recording";
+            recordButton.classList.remove("recording");
+            clearInterval(volumeChecker);
         }
 
-        function toggleRecording() {
+        function sendAudioToServer(data) {
-            if (recording) {
+            if (connected) {
-                stopRecording();
+                //const blob = new Blob(data, { type: 'audio/webm' });
+                socket.send(data);
+                serverTime = Date.now();
+                console.log("Sent some audio data to server.");
+                if (!autosend.checked) {
+                    transcription.innerHTML = "Processing audio...";
+                }
             } else {
-                startRecording();
+                console.log("Not connected, not sending audio data to server.");
+            }
+        }
+        function toggleListening() {
+            if (socket.readyState === WebSocket.OPEN) {
+                if (recording) {
+                    stopListening();
+                } else {
+                    startListening();
+                }
             }
         }
 
         window.onload = () => {
             recordButton = document.getElementById("record-button");
-            recordButton.addEventListener('click', toggleRecording);
+            recordButton.addEventListener("click", toggleListening);
             connectionStatus = document.getElementById("connection-status");
+            //transcription = document.getElementById("transcription");
+            //autosend = document.getElementById("autosend");
+            statusRecording = document.getElementById("status-recording");
+
+
+
             connect(socket);
         };
     </script>
     <script src="https://cdn.webrtc-experiment.com/MediaStreamRecorder.js"></script>
-  </body>
+</body>
+
 </html>

web/server.js

@@ -2,17 +2,27 @@ const WebSocket = require('ws');
 const wss = new WebSocket.Server({ port: 8081 });
 console.log('WebSocket server started on port 8081');
 
+
+
+//load TTS_BACHEND_URL from .env file
+//require('dotenv').config();
+
+console.log(process.env)
+console.log(process.env.TTS_BACKEND_URL)
+console.log(process.env.WS_URL)
+
 //we use https://hub.docker.com/r/onerahmet/openai-whisper-asr-webservice to transcribe the audio
 //docker run -p 9009:9009 -d onerahmet/openai-whisper-asr-webservice
 
 wss.on('connection', (ws) => {
     console.log('Client ' + ws._socket.remoteAddress + ' connected');
     ws.on('message', (data) => {
-        console.log('Received data from client: ' + data.length + ' bytes');
+        //show the size of the audio data as 0.000 MB
+        console.log('Received data from client: ' + (data.length / 1024 / 1024).toFixed(3) + ' MB');
         var request = require('request');
         var formData = {
             task: 'transcribe',
-            language: 'en-US',
+            language: 'en-US', //bg-BG|en-US
             output: 'json',
             audio_file: {
                 value: data,
@@ -22,15 +32,28 @@ wss.on('connection', (ws) => {
                 }
             }
         };
-        request.post({url:'http://192.168.0.10:9009/asr', formData: formData}, function optionalCallback(err, httpResponse, body) {
+        //save the audio data to a file to /rec folder
+        var fs = require('fs');
+        var timestampfilename = Date.now();
+        fs.writeFile('./rec/audio' + timestampfilename + '.ogg', data, function (err) {
+
+        
+        //fs.writeFile('audio' + timestampfilename + '.ogg', data, function (err) {
+            if (err) {
+                return console.log(err);
+            }
+            console.log('Audio data saved to audio.ogg');
+        });
+
+
+
+        request.post({url:'http://192.168.0.10:9008/asr', formData: formData}, function optionalCallback(err, httpResponse, body) {
             if (err) {
                 return console.error('upload failed:', err);
             }
-            console.log('Upload successful!  Server responded with:', body);
+            console.log('Whisper decoded:', body);
-            ws.send(">>: " + body);
+            ws.send(body);
         });
-
-        ws.send("Processing audio...");
     });
 });
 
@@ -50,6 +73,13 @@ app.get('/', (req, res) => {
     res.sendFile(path.join(__dirname, 'client.html'));
 });
 
+
+//get WS url from .env file
+app.get('/wsurl', (req, res) => {
+    res.send(process.env.WS_URL, 200, {'Content-Type': 'text/plain'});
+});
+
+
 app.listen(8080, () => {
     console.log('Server listening on port 8080');
 });