mirror of
https://github.com/SrIzan10/vdo.ninja.git
synced 2026-05-01 11:05:24 +00:00
247 lines
10 KiB
HTML
247 lines
10 KiB
HTML
<!DOCTYPE html>
|
|
<html lang="en">
|
|
<head>
|
|
<meta charset="UTF-8">
|
|
<meta name="viewport" content="width=device-width, initial-scale=1.0">
|
|
<title>Adaptive Loudness Transition Analysis</title>
|
|
<style>
|
|
body {
|
|
font-family: Arial, sans-serif;
|
|
line-height: 1.6;
|
|
color: #333;
|
|
max-width: 800px;
|
|
margin: 0 auto;
|
|
padding: 20px;
|
|
background-color: #f4f4f4;
|
|
}
|
|
h1, h2 {
|
|
color: #2c3e50;
|
|
text-align: center;
|
|
}
|
|
.input-group {
|
|
margin-bottom: 20px;
|
|
background-color: #fff;
|
|
padding: 20px;
|
|
border-radius: 5px;
|
|
box-shadow: 0 2px 5px rgba(0,0,0,0.1);
|
|
}
|
|
label {
|
|
display: block;
|
|
margin-bottom: 5px;
|
|
}
|
|
input[type="file"], input[type="number"] {
|
|
width: 100%;
|
|
padding: 10px;
|
|
margin-bottom: 10px;
|
|
border: 1px solid #ddd;
|
|
border-radius: 4px;
|
|
}
|
|
#error {
|
|
color: #e74c3c;
|
|
font-weight: bold;
|
|
}
|
|
#results, #debug {
|
|
background-color: #fff;
|
|
padding: 20px;
|
|
border-radius: 5px;
|
|
box-shadow: 0 2px 5px rgba(0,0,0,0.1);
|
|
margin-top: 20px;
|
|
}
|
|
.region {
|
|
margin-bottom: 10px;
|
|
padding: 10px;
|
|
background-color: #ecf0f1;
|
|
border-radius: 4px;
|
|
}
|
|
.region-header {
|
|
font-weight: bold;
|
|
color: #2980b9;
|
|
}
|
|
#debug {
|
|
font-family: monospace;
|
|
white-space: pre-wrap;
|
|
background-color: #2c3e50;
|
|
color: #ecf0f1;
|
|
}
|
|
</style>
|
|
</head>
|
|
<body>
|
|
<h1>Adaptive Loudness Transition Analysis</h1>
|
|
<div class="input-group">
|
|
<label for="audioFile">Select WAV file:</label>
|
|
<input type="file" id="audioFile" accept=".wav">
|
|
<label for="transitionThreshold">Transition Threshold (dB):</label>
|
|
<input type="number" id="transitionThreshold" value="5" min="1" max="20" step="0.5">
|
|
<label for="analysisWindowSize">Analysis Window Size (seconds):</label>
|
|
<input type="number" id="analysisWindowSize" value="5" min="1" max="30" step="1">
|
|
</div>
|
|
<div id="error"></div>
|
|
<div id="results"></div>
|
|
<div id="debug"></div>
|
|
|
|
<script>
|
|
let audioContext;
|
|
|
|
document.getElementById('audioFile').addEventListener('change', handleFileUpload);
|
|
|
|
async function handleFileUpload(event) {
|
|
const errorDiv = document.getElementById('error');
|
|
const resultsDiv = document.getElementById('results');
|
|
const debugDiv = document.getElementById('debug');
|
|
errorDiv.innerHTML = '';
|
|
resultsDiv.innerHTML = '';
|
|
debugDiv.innerHTML = '';
|
|
|
|
try {
|
|
audioContext = new (window.AudioContext || window.webkitAudioContext)();
|
|
|
|
const file = event.target.files[0];
|
|
if (!file) {
|
|
throw new Error('No file selected');
|
|
}
|
|
|
|
const arrayBuffer = await file.arrayBuffer();
|
|
const audioBuffer = await audioContext.decodeAudioData(arrayBuffer);
|
|
|
|
console.log('Audio duration:', audioBuffer.duration, 'seconds');
|
|
debugDiv.innerHTML += `Audio duration: ${audioBuffer.duration.toFixed(2)} seconds\n`;
|
|
|
|
const loudnessData = analyzeLoudness(audioBuffer);
|
|
console.log('Loudness data:', loudnessData);
|
|
debugDiv.innerHTML += `Loudness data length: ${loudnessData.length}\n`;
|
|
debugDiv.innerHTML += `First 5 loudness data points: ${JSON.stringify(loudnessData.slice(0, 5), null, 2)}\n`;
|
|
debugDiv.innerHTML += `Last 5 loudness data points: ${JSON.stringify(loudnessData.slice(-5), null, 2)}\n`;
|
|
|
|
if (loudnessData.length === 0) {
|
|
throw new Error('No loudness data generated. The audio file might be too short.');
|
|
}
|
|
|
|
const silenceThreshold = determineSilenceThreshold(loudnessData);
|
|
debugDiv.innerHTML += `Automatically determined silence threshold: ${silenceThreshold.toFixed(2)} dB\n`;
|
|
|
|
const transitionThreshold = parseFloat(document.getElementById('transitionThreshold').value);
|
|
const analysisWindowSize = parseFloat(document.getElementById('analysisWindowSize').value);
|
|
const regions = identifyLoudnessTransitions(loudnessData, silenceThreshold, transitionThreshold, analysisWindowSize);
|
|
console.log('Regions:', regions);
|
|
debugDiv.innerHTML += `Number of regions identified: ${regions.length}\n`;
|
|
debugDiv.innerHTML += `Regions: ${JSON.stringify(regions, null, 2)}\n`;
|
|
|
|
displayResults(regions, audioBuffer.duration, silenceThreshold);
|
|
} catch (error) {
|
|
console.error('Error:', error);
|
|
errorDiv.innerHTML = `Error: ${error.message}`;
|
|
debugDiv.innerHTML += `Error stack trace: ${error.stack}\n`;
|
|
}
|
|
}
|
|
|
|
function analyzeLoudness(audioBuffer) {
|
|
const channelData = audioBuffer.getChannelData(0);
|
|
const sampleRate = audioBuffer.sampleRate;
|
|
const segmentLength = 0.1 * sampleRate; // 100ms segments
|
|
const loudnessData = [];
|
|
|
|
for (let i = 0; i < channelData.length; i += segmentLength) {
|
|
const segment = channelData.slice(i, i + segmentLength);
|
|
const loudness = calculateLoudness(segment);
|
|
if (!isNaN(loudness)) {
|
|
loudnessData.push({ time: i / sampleRate, loudness });
|
|
}
|
|
}
|
|
|
|
return loudnessData;
|
|
}
|
|
|
|
function calculateLoudness(samples) {
|
|
if (samples.length === 0) return NaN;
|
|
const rms = Math.sqrt(samples.reduce((sum, sample) => sum + sample * sample, 0) / samples.length);
|
|
return 20 * Math.log10(Math.max(rms, 1e-10)); // Avoid log of 0
|
|
}
|
|
|
|
function determineSilenceThreshold(loudnessData) {
|
|
const sortedLoudness = loudnessData.map(d => d.loudness).sort((a, b) => a - b);
|
|
const medianIndex = Math.floor(sortedLoudness.length / 2);
|
|
const median = sortedLoudness[medianIndex];
|
|
|
|
// Assuming half of the audio should be silence, we'll use the median as our threshold
|
|
return median;
|
|
}
|
|
|
|
function identifyLoudnessTransitions(loudnessData, silenceThreshold, transitionThreshold, analysisWindowSize) {
|
|
const windowSamples = Math.floor(analysisWindowSize / 0.1); // Convert seconds to number of 100ms samples
|
|
let regions = [];
|
|
let currentRegion = null;
|
|
let prevAvgLoudness = null;
|
|
|
|
for (let i = 0; i < loudnessData.length - windowSamples; i += windowSamples) {
|
|
const window = loudnessData.slice(i, i + windowSamples);
|
|
const avgLoudness = average(window.map(d => d.loudness));
|
|
|
|
if (avgLoudness > silenceThreshold) {
|
|
if (!currentRegion) {
|
|
currentRegion = { start: window[0].time, avgLoudness };
|
|
} else if (prevAvgLoudness !== null && Math.abs(avgLoudness - prevAvgLoudness) > transitionThreshold) {
|
|
currentRegion.end = window[0].time;
|
|
regions.push(currentRegion);
|
|
currentRegion = { start: window[0].time, avgLoudness };
|
|
}
|
|
prevAvgLoudness = avgLoudness;
|
|
} else if (currentRegion) {
|
|
currentRegion.end = window[0].time;
|
|
regions.push(currentRegion);
|
|
currentRegion = null;
|
|
prevAvgLoudness = null;
|
|
}
|
|
}
|
|
|
|
if (currentRegion) {
|
|
currentRegion.end = loudnessData[loudnessData.length - 1].time;
|
|
regions.push(currentRegion);
|
|
}
|
|
|
|
return regions;
|
|
}
|
|
|
|
function average(arr) {
|
|
return arr.reduce((sum, val) => sum + val, 0) / arr.length;
|
|
}
|
|
|
|
function displayResults(regions, totalDuration, silenceThreshold) {
|
|
const resultsDiv = document.getElementById('results');
|
|
resultsDiv.innerHTML = '<h2>Analysis Results</h2>';
|
|
resultsDiv.innerHTML += `<p>Automatically determined silence threshold: ${silenceThreshold.toFixed(2)} dB</p>`;
|
|
if (regions.length === 0) {
|
|
resultsDiv.innerHTML += '<p>No significant loudness transitions identified. Try adjusting the parameters.</p>';
|
|
return;
|
|
}
|
|
regions.forEach((region, i) => {
|
|
const regionDuration = (region.end - region.start).toFixed(2);
|
|
const percentOfTotal = ((region.end - region.start) / totalDuration * 100).toFixed(1);
|
|
const loudnessDescription = describeLoudness(region.avgLoudness);
|
|
resultsDiv.innerHTML += `
|
|
<div class="region">
|
|
<span class="region-header">Region ${i + 1}:</span><br>
|
|
Start: ${formatTime(region.start)}<br>
|
|
End: ${formatTime(region.end)}<br>
|
|
Duration: ${regionDuration}s (${percentOfTotal}% of total)<br>
|
|
Avg Loudness: ${region.avgLoudness.toFixed(2)} dB (${loudnessDescription})
|
|
</div>
|
|
`;
|
|
});
|
|
}
|
|
|
|
function formatTime(seconds) {
|
|
const minutes = Math.floor(seconds / 60);
|
|
const remainingSeconds = (seconds % 60).toFixed(2);
|
|
return `${minutes}:${remainingSeconds.padStart(5, '0')}`;
|
|
}
|
|
|
|
function describeLoudness(loudness) {
|
|
if (loudness > -10) return "Very Loud";
|
|
if (loudness > -20) return "Loud";
|
|
if (loudness > -30) return "Moderate";
|
|
if (loudness > -40) return "Soft";
|
|
return "Very Soft";
|
|
}
|
|
</script>
|
|
</body>
|
|
</html> |