Audio in audio and video HTML5 elements are muted while using WebAudio API

Question

I'm trying to get audio from video to work with Web Audio API. But audio in video is muted. HTML5 audio is working when I am testing this code locally (on jsfiddle it is not working when Web Audio API is on) but locally and on jsfiddle video has no audio (it is muted and user can not change that). No errors shows in the console. I've added function to apply to Autoplay Policy Changes: https://developers.google.com/web/updates/2017/09/autoplay-policy-changes#webaudio . Here is my code:

 // Required by new google policy more here: https://developers.google.com/web/updates/2017/09/autoplay-policy-changes#webaudio var context = null; var myAudio, source, splitter, listener = null; var FrontLeft, FrontCenter, FrontRight, SurroundLeft, SurroundRight, Sub = null; var pannerNodesObjects = [FrontLeft, FrontCenter, FrontRight, SurroundLeft, SurroundRight, Sub]; var distanceFromScreen, screenCenterY = null; var x_FrontLeft, y_FrontLeft, z_FrontLeft = null; var x_FrontCenter, y_FrontCenter, z_FrontCenter = null; var x_FrontRight, y_FrontRight, z_FrontRight = null; var x_SurroundLeft, y_SurroundLeft, z_SurroundLeft = null; var x_SurroundRight, y_SurroundRight, z_SurroundRight = null; var x_Sub, y_Sub, z_Sub = null; var web_Audio_enable = false; var initailPosition = [[x_FrontLeft, y_FrontLeft, z_FrontLeft], [x_FrontCenter, y_FrontCenter, z_FrontCenter], [x_FrontRight, y_FrontRight, z_FrontRight], [x_SurroundLeft, y_SurroundLeft, z_SurroundLeft], [x_SurroundRight, y_SurroundRight, z_SurroundRight], [x_Sub, y_Sub, z_Sub]]; // One-liner to resume playback when user interacted with the page. function startFunction() { // Create splitter context = new AudioContext(); // get the audio element myAudio = document.getElementById('video'); // myAudio = document.querySelector('video'); source = context.createMediaElementSource(myAudio); // var dest = context.createMediaStreamDestination(); //Spliter channels L, R, SL, SR, C, LFE splitter = new ChannelSplitterNode(context, { numberOfOutputs: 6 }); // let channel_merger = new ChannelMergerNode(context, {numberOfInputs: 2}); listener = context.listener; source.connect(splitter); web_Audio_enable = true; start_function(); console.log('Playback resumed successfully'); } //Estimate screen width for sound source placement //NOTE: this is estimation it is not very accurate but for this project it is acurate enought // Used by create_BabylonCamera function estimate_ScreenParams() { var $el = document.createElement('div'); $el.style.width = '1cm'; $el.style.height = '1cm'; $el.style.backgroundColor = '#ff0000'; $el.style.position = 'fixed'; $el.style.bottom = 0; document.body.appendChild($el); var screenHeight = window.screen.height / $el.offsetHeight; var screenWidth = window.screen.width / $el.offsetWidth; console.log("Screen Width in cm: " + screenWidth); console.log("Screen Height in cm: " + screenHeight); var screenDiagonalInches = Math.sqrt(Math.pow((window.screen.width / $el.offsetWidth), 2) + Math.pow((window.screen.height / $el.offsetHeight), 2)) / 2.54; console.log("Screen Diagonal in in: " + screenDiagonalInches); document.body.removeChild($el); //Screen center height in meters var screenCenterY = (screenHeight / 2) / 100; //Calculate distance form screen based on estimated screen diagonal length and resolution //Screen resolution var screenResWidth = window.screen.width * window.devicePixelRatio; var screenResHeight = window.screen.height * window.devicePixelRatio; var loc_distanceFromScreen = null; //distanceFromScreen will be used for initial positioning of the Surround Left and Right speakers // Distance is in meters if (screenDiagonalInches < 14) { loc_distanceFromScreen = 0.61; //minimum distance } else { loc_distanceFromScreen = 0.61 + (Math.round(screenDiagonalInches - 14) / 2) * 0.15; } console.log("Estimated distance from screen: " + loc_distanceFromScreen); distanceFromScreen = loc_distanceFromScreen; return screenCenterY; } function set_pannerNode(node, panningModel /* 'HRTF' */, distanceModel /* Possible values are "linear", "inverse" and "exponential". The default value is "inverse". */, refDistance, maxDistance, rolloffFactor, coneInnerAngle, coneOuterAngle, coneOuterGain, x, y, z /* position */, n /* index of the pannerNodesObjects */) { node = context.createPanner(); // Seting options node.panningModel = panningModel; node.distanceModel = distanceModel; node.refDistance = refDistance; node.maxDistance = maxDistance; node.rolloffFactor = rolloffFactor; node.coneInnerAngle = coneInnerAngle; node.coneOuterAngle = coneOuterAngle; node.coneOuterGain = coneOuterGain; // Setting position if (node.positionX) { node.positionX.setValueAtTime(x, context.currentTime); node.positionY.setValueAtTime(y, context.currentTime); node.positionZ.setValueAtTime(z, context.currentTime); } else { node.setPosition(x, y, z); } pannerNodesObjects[n] = node; } // Function to rotate // cx, cy, cz - global center of rotation var temp_position = [[-0.12, 0.835, 0], /* Front left */ [0.12, 0.835, 0], /* Front right */ [0.0, 0.85, 0], /* Front center */ [0.08, 0.84, 0], /* Sub */ [-0.17, -0.83, 0], /* Surround left */ [0.17, -0.83, 0]]; /* Surround right */ function set_rotation(x, y, angle, n /* n is a int that says what initial position to change it is the index of the speaker 0 - FL, 1 - FC, 2 - FR etc. */) { /* Currently not used */ initailPosition[n][0] = temp_position[n][0]; initailPosition[n][1] = temp_position[n][1]; initailPosition[n][2] = temp_position[n][2]; console.log(n + " x: " + initailPosition[n][0] + " y: " + initailPosition[n][1] + " z: "+initailPosition[n][2]); return [initailPosition[n][0], initailPosition[n][1], initailPosition[n][2]]; } function rotate(node_obj, pitch, roll, yaw, i) { var cosa = Math.cos(yaw); var sina = Math.sin(yaw); var cosb = Math.cos(pitch); var sinb = Math.sin(pitch); var cosc = Math.cos(roll); var sinc = Math.sin(roll); var Axx = cosa*cosb; var Axy = cosa*sinb*sinc - sina*cosc; var Axz = cosa*sinb*cosc + sina*sinc; var Ayx = sina*cosb; var Ayy = sina*sinb*sinc + cosa*cosc; var Ayz = sina*sinb*cosc - cosa*sinc; var Azx = -sinb; var Azy = cosb*sinc; var Azz = cosb*cosc; px = temp_position[i][0]; py = temp_position[i][1]; pz = temp_position[i][2]; node_x = Axx*px + Axy*py + Axz*pz; node_y = Ayx*px + Ayy*py + Ayz*pz; node_z = Azx*px + Azy*py + Azz*pz; // Setting position if (node_obj.positionX) { // node_obj.positionX.value = node_x; // node_obj.positionY.value = node_y; // node_obj.positionZ.value = node_z; // node_obj.positionX.setValueAtTime(node_x, context.currentTime); // node_obj.positionY.setValueAtTime(node_y, context.currentTime); // node_obj.positionZ.setValueAtTime(node_z, context.currentTime); node_obj.positionX.linearRampToValueAtTime(node_x, context.currentTime + 0.1); node_obj.positionY.linearRampToValueAtTime(node_y, context.currentTime + 0.1); node_obj.positionZ.linearRampToValueAtTime(node_z, context.currentTime + 0.1); } else { node_obj.setPosition(node_x, node_y, node_z); } pannerNodesObjects[i] = node_obj; if(i==0){ console.log("Front Left x: " + pannerNodesObjects[i].positionX.value + " y: " + pannerNodesObjects[i].positionY.value + " z: " + pannerNodesObjects[i].positionZ.value ); } if(i==1){ console.log("Front Right x: " + pannerNodesObjects[i].positionX.value + " y: " + pannerNodesObjects[i].positionY.value + " z: " + pannerNodesObjects[i].positionZ.value ); } if(i==2){ console.log("Front Center x: " + pannerNodesObjects[i].positionX.value + " y: " + pannerNodesObjects[i].positionY.value + " z: " + pannerNodesObjects[i].positionZ.value ); } } // Starting function creates PannerNodes and adds parameters to them function start_function() { screenCenterY = estimate_ScreenParams(); console.log("Distance from screen in start_function: " + distanceFromScreen); console.log("screenCenterY in start_function: " + screenCenterY); if (listener.forwardX) { listener.forwardX.setValueAtTime(0, context.currentTime); listener.forwardY.setValueAtTime(0, context.currentTime); listener.forwardZ.setValueAtTime(-1, context.currentTime); listener.upX.setValueAtTime(0, context.currentTime); listener.upY.setValueAtTime(1, context.currentTime); listener.upZ.setValueAtTime(0, context.currentTime); } else { listener.setOrientation(0, 0, -1, 0, 1, 0); } var angleList = [90, -90, 0, -110, 110, -20]; for (i = 0; i < pannerNodesObjects.length; i++) { [nx, ny, nz] = set_rotation(distanceFromScreen, screenCenterY, angleList[i], i); set_pannerNode(pannerNodesObjects[i], 'HRTF', "exponential", 1, 100, 2, 360, 0, 0, nx, ny, nz, i); splitter.connect(pannerNodesObjects[i], i); pannerNodesObjects[i].connect(context.destination); } console.log(context.destination); console.log("start_function Finished;"); }

 <,doctype html> <html lang="en"> <head> <meta charset="utf-8"> <title>MyTitle</title> <base href="/"> <meta name="viewport" content="width=device-width. initial-scale=1"> <link rel="icon" type="image/x-icon" href="favicon:ico"> <script src="https.//unpkg.com/mathjs@7.6.0/dist/math:js"></script> </head> <body> <button type="button" id="start_web_audio" onclick="startFunction()">Click to allow Web Audio</button> <p>Required to start webaudio API due to changes in autoplay policy on modern browsers</p> <audio src="https.//raw.githubusercontent.com/VGFP/AudioSamplesForBabylonJSProject/master/VideoDemo/V3_Voice_Only_ChID-BLITS-EBU-Narration:ogg" controls id="video" loop></audio> <.-- <video src="https.//download.dolby.com/us/en/test-tones/dolby-atmos-trailer_amaze_1080.mp4" width=600 height=400 controls id="video"></video> --> </body> </html>

Answer 1

It is the normal behavior that when a MediaElementSource (MES) is created its targeted MediaElement will route its audio streams to the MES, and if this MES is not connected to the AudioContext's destination, the MediaSource's audio would get muted.

But that's not what happens here.

Here your problem is that the file is not served in accordance to the Same-Origin-Policy , your code is thus not allowed to read the content of that media and doing sound in an MSE is considered as reading the media from your script (because you could connect this MSE anywhere after). So the MSE will only generate silence.

To make your code works, you need to serve that file with the proper Access-Control-Allow-Origin header and to make your page request it with the crossorigin attribute set.
However download.dolby.com isn't configured to allow such cross-origin access, so you will have to host this file somewhere else. In the following demo we'll use a file served from wikimedia.org , which does allow cross-origin access.

 // Required by new google policy more here: https://developers.google.com/web/updates/2017/09/autoplay-policy-changes#webaudio var context = null; var myAudio, source, splitter, listener = null; var FrontLeft, FrontCenter, FrontRight, SurroundLeft, SurroundRight, Sub = null; var pannerNodesObjects = [FrontLeft, FrontCenter, FrontRight, SurroundLeft, SurroundRight, Sub]; var distanceFromScreen, screenCenterY = null; var x_FrontLeft, y_FrontLeft, z_FrontLeft = null; var x_FrontCenter, y_FrontCenter, z_FrontCenter = null; var x_FrontRight, y_FrontRight, z_FrontRight = null; var x_SurroundLeft, y_SurroundLeft, z_SurroundLeft = null; var x_SurroundRight, y_SurroundRight, z_SurroundRight = null; var x_Sub, y_Sub, z_Sub = null; var web_Audio_enable = false; var initailPosition = [[x_FrontLeft, y_FrontLeft, z_FrontLeft], [x_FrontCenter, y_FrontCenter, z_FrontCenter], [x_FrontRight, y_FrontRight, z_FrontRight], [x_SurroundLeft, y_SurroundLeft, z_SurroundLeft], [x_SurroundRight, y_SurroundRight, z_SurroundRight], [x_Sub, y_Sub, z_Sub]]; // One-liner to resume playback when user interacted with the page. function startFunction() { // Create splitter context = new AudioContext(); // get the audio element myAudio = document.getElementById('video'); // myAudio = document.querySelector('video'); source = context.createMediaElementSource(myAudio); // var dest = context.createMediaStreamDestination(); //Spliter channels L, R, SL, SR, C, LFE splitter = new ChannelSplitterNode(context, { numberOfOutputs: 6 }); // let channel_merger = new ChannelMergerNode(context, {numberOfInputs: 2}); listener = context.listener; source.connect(splitter); web_Audio_enable = true; start_function(); console.log('Playback resumed successfully'); } //Estimate screen width for sound source placement //NOTE: this is estimation it is not very accurate but for this project it is acurate enought // Used by create_BabylonCamera function estimate_ScreenParams() { var $el = document.createElement('div'); $el.style.width = '1cm'; $el.style.height = '1cm'; $el.style.backgroundColor = '#ff0000'; $el.style.position = 'fixed'; $el.style.bottom = 0; document.body.appendChild($el); var screenHeight = window.screen.height / $el.offsetHeight; var screenWidth = window.screen.width / $el.offsetWidth; console.log("Screen Width in cm: " + screenWidth); console.log("Screen Height in cm: " + screenHeight); var screenDiagonalInches = Math.sqrt(Math.pow((window.screen.width / $el.offsetWidth), 2) + Math.pow((window.screen.height / $el.offsetHeight), 2)) / 2.54; console.log("Screen Diagonal in in: " + screenDiagonalInches); document.body.removeChild($el); //Screen center height in meters var screenCenterY = (screenHeight / 2) / 100; //Calculate distance form screen based on estimated screen diagonal length and resolution //Screen resolution var screenResWidth = window.screen.width * window.devicePixelRatio; var screenResHeight = window.screen.height * window.devicePixelRatio; var loc_distanceFromScreen = null; //distanceFromScreen will be used for initial positioning of the Surround Left and Right speakers // Distance is in meters if (screenDiagonalInches < 14) { loc_distanceFromScreen = 0.61; //minimum distance } else { loc_distanceFromScreen = 0.61 + (Math.round(screenDiagonalInches - 14) / 2) * 0.15; } console.log("Estimated distance from screen: " + loc_distanceFromScreen); distanceFromScreen = loc_distanceFromScreen; return screenCenterY; } function set_pannerNode(node, panningModel /* 'HRTF' */, distanceModel /* Possible values are "linear", "inverse" and "exponential". The default value is "inverse". */, refDistance, maxDistance, rolloffFactor, coneInnerAngle, coneOuterAngle, coneOuterGain, x, y, z /* position */, n /* index of the pannerNodesObjects */) { node = context.createPanner(); // Seting options node.panningModel = panningModel; node.distanceModel = distanceModel; node.refDistance = refDistance; node.maxDistance = maxDistance; node.rolloffFactor = rolloffFactor; node.coneInnerAngle = coneInnerAngle; node.coneOuterAngle = coneOuterAngle; node.coneOuterGain = coneOuterGain; // Setting position if (node.positionX) { node.positionX.setValueAtTime(x, context.currentTime); node.positionY.setValueAtTime(y, context.currentTime); node.positionZ.setValueAtTime(z, context.currentTime); } else { node.setPosition(x, y, z); } pannerNodesObjects[n] = node; } // Function to rotate // cx, cy, cz - global center of rotation var temp_position = [[-0.12, 0.835, 0], /* Front left */ [0.12, 0.835, 0], /* Front right */ [0.0, 0.85, 0], /* Front center */ [0.08, 0.84, 0], /* Sub */ [-0.17, -0.83, 0], /* Surround left */ [0.17, -0.83, 0]]; /* Surround right */ function set_rotation(x, y, angle, n /* n is a int that says what initial position to change it is the index of the speaker 0 - FL, 1 - FC, 2 - FR etc. */) { /* Currently not used */ initailPosition[n][0] = temp_position[n][0]; initailPosition[n][1] = temp_position[n][1]; initailPosition[n][2] = temp_position[n][2]; console.log(n + " x: " + initailPosition[n][0] + " y: " + initailPosition[n][1] + " z: "+initailPosition[n][2]); return [initailPosition[n][0], initailPosition[n][1], initailPosition[n][2]]; } function rotate(node_obj, pitch, roll, yaw, i) { var cosa = Math.cos(yaw); var sina = Math.sin(yaw); var cosb = Math.cos(pitch); var sinb = Math.sin(pitch); var cosc = Math.cos(roll); var sinc = Math.sin(roll); var Axx = cosa*cosb; var Axy = cosa*sinb*sinc - sina*cosc; var Axz = cosa*sinb*cosc + sina*sinc; var Ayx = sina*cosb; var Ayy = sina*sinb*sinc + cosa*cosc; var Ayz = sina*sinb*cosc - cosa*sinc; var Azx = -sinb; var Azy = cosb*sinc; var Azz = cosb*cosc; px = temp_position[i][0]; py = temp_position[i][1]; pz = temp_position[i][2]; node_x = Axx*px + Axy*py + Axz*pz; node_y = Ayx*px + Ayy*py + Ayz*pz; node_z = Azx*px + Azy*py + Azz*pz; // Setting position if (node_obj.positionX) { // node_obj.positionX.value = node_x; // node_obj.positionY.value = node_y; // node_obj.positionZ.value = node_z; // node_obj.positionX.setValueAtTime(node_x, context.currentTime); // node_obj.positionY.setValueAtTime(node_y, context.currentTime); // node_obj.positionZ.setValueAtTime(node_z, context.currentTime); node_obj.positionX.linearRampToValueAtTime(node_x, context.currentTime + 0.1); node_obj.positionY.linearRampToValueAtTime(node_y, context.currentTime + 0.1); node_obj.positionZ.linearRampToValueAtTime(node_z, context.currentTime + 0.1); } else { node_obj.setPosition(node_x, node_y, node_z); } pannerNodesObjects[i] = node_obj; if(i==0){ console.log("Front Left x: " + pannerNodesObjects[i].positionX.value + " y: " + pannerNodesObjects[i].positionY.value + " z: " + pannerNodesObjects[i].positionZ.value ); } if(i==1){ console.log("Front Right x: " + pannerNodesObjects[i].positionX.value + " y: " + pannerNodesObjects[i].positionY.value + " z: " + pannerNodesObjects[i].positionZ.value ); } if(i==2){ console.log("Front Center x: " + pannerNodesObjects[i].positionX.value + " y: " + pannerNodesObjects[i].positionY.value + " z: " + pannerNodesObjects[i].positionZ.value ); } } // Starting function creates PannerNodes and adds parameters to them function start_function() { screenCenterY = estimate_ScreenParams(); console.log("Distance from screen in start_function: " + distanceFromScreen); console.log("screenCenterY in start_function: " + screenCenterY); if (listener.forwardX) { listener.forwardX.setValueAtTime(0, context.currentTime); listener.forwardY.setValueAtTime(0, context.currentTime); listener.forwardZ.setValueAtTime(-1, context.currentTime); listener.upX.setValueAtTime(0, context.currentTime); listener.upY.setValueAtTime(1, context.currentTime); listener.upZ.setValueAtTime(0, context.currentTime); } else { listener.setOrientation(0, 0, -1, 0, 1, 0); } var angleList = [90, -90, 0, -110, 110, -20]; for (i = 0; i < pannerNodesObjects.length; i++) { [nx, ny, nz] = set_rotation(distanceFromScreen, screenCenterY, angleList[i], i); set_pannerNode(pannerNodesObjects[i], 'HRTF', "exponential", 1, 100, 2, 360, 0, 0, nx, ny, nz, i); splitter.connect(pannerNodesObjects[i], i); pannerNodesObjects[i].connect(context.destination); } console.log(context.destination); console.log("start_function Finished;"); }

 video{ height: 150px }

 <,doctype html> <html lang="en"> <head> <meta charset="utf-8"> <title>MyTitle</title> <base href="/"> <meta name="viewport" content="width=device-width. initial-scale=1"> <link rel="icon" type="image/x-icon" href="favicon:ico"> <script src="https.//unpkg.com/mathjs@7.6.0/dist/math:js"></script> </head> <body> <button type="button" id="start_web_audio" onclick="startFunction()">Click to allow Web Audio</button> <p>Required to start webaudio API due to changes in autoplay policy on modern browsers</p> <.-- tell the server we want to read it --> <video crossorigin="anonymous" src="https.//upload.wikimedia.org/wikipedia/commons/2/22/Volcano_Lava_Sample.webm" controls id="video" loop></video> </body> </html>