ajax/games
1
0
Fork 0
mirror of https://github.com/hexahigh/games.git synced 2025-12-11 20:15:38 +01:00
Code Issues Packages Projects Releases Wiki Activity

updated ai model

Browse Source
This commit is contained in:
Boof 2023-09-11 12:25:24 +02:00 committed by GitHub
parent 2fe5b26692
commit 12588fc65f
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 62 additions and 45 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
other/tensor/test1/index.html
View File

@ -3,13 +3,15 @@
<head> <head>
<title>Audio Classification</title> <title>Audio Classification</title>
<script src="https://cdn.jsdelivr.net/npm/@tensorflow/tfjs@2.8.4/dist/tf.min.js"></script> <script src="https://cdn.jsdelivr.net/npm/@tensorflow/tfjs@2.8.4/dist/tf.min.js"></script>
<script src="https://cdn.jsdelivr.net/npm/@tensorflow-models/speech-commands@0.4.0/dist/speech-commands.min.js"></script>
<script src="script.js" defer></script> <script src="script.js" defer></script>
</head> </head>
<body> <body>
<div>Audio Model</div> <h1>Audio Classification</h1>
<button type="button" onclick="init()">Start</button> <form>
<div id="label-container"></div> <input type="file" id="audioFile" name="audioFile"/>
<button type="submit">Upload</button>
</form>
<section id="result"></section>
</body> </body>
</html> </html>

95
other/tensor/test1/script.js
View File

@ -1,48 +1,63 @@
// more documentation available at // Load the TensorFlow.js model
// https://github.com/tensorflow/tfjs-models/tree/master/speech-commands async function loadModel() {
const model = await tf.loadLayersModel("model/model.json");
return model;
}
// the link to your model provided by Teachable Machine export panel function preprocessAudio(filename) {
const URL = "./model/"; // Load audio file
const audio = new Audio(filename);
async function createModel() { // Generate a Mel-scaled spectrogram
const checkpointURL = URL + "model.json"; // model topology const spectrogram = tf.tidy(() => {
const audioTensor = tf.browser.fromAudio(audio);
const sampleRate = audioTensor.sampleRate();
const melSpectrogram = tf.signal.melSpectrogram(audioTensor, {
sampleRate: sampleRate,
hopLength: 512,
nMels: 128,
fMin: 0,
fMax: 8000
});
const logMelSpectrogram = tf.math.log(melSpectrogram.add(1e-6));
const normalizedSpectrogram = logMelSpectrogram.div(tf.max(logMelSpectrogram));
return normalizedSpectrogram;
});
const recognizer = speechCommands.create( return spectrogram;
"BROWSER_FFT", // fourier transform type, not useful to change }
undefined, // speech commands vocabulary feature, not useful for your models
checkpointURL);
// check that model and metadata are loaded via HTTPS requests.
await recognizer.ensureModelLoaded();
return recognizer; // Classify the audio using the loaded model
} async function classifyAudio(model, audioData) {
const preprocessedData = preprocessAudio(audioData);
const audioTensor = tf.tensor(preprocessedData);
const predictions = model.predict(audioTensor);
const predictedClass = predictions.argMax().dataSync()[0];
return predictedClass;
}
async function init() { // Handle form submission
const recognizer = await createModel(); document.querySelector("form").addEventListener("submit", async function(event) {
const classLabels = recognizer.wordLabels(); // get class labels event.preventDefault();
const labelContainer = document.getElementById("label-container"); const fileInput = document.getElementById("audioFile");
for (let i = 0; i < classLabels.length; i++) { const file = fileInput.files[0];
labelContainer.appendChild(document.createElement("div")); const fileReader = new FileReader();
}
// listen() takes two arguments: fileReader.onload = async function(event) {
// 1. A callback function that is invoked anytime a word is recognized. const audioData = event.target.result;
// 2. A configuration object with adjustable fields
recognizer.listen(result => {
const scores = result.scores; // probability of prediction for each class
// render the probability scores per class
for (let i = 0; i < classLabels.length; i++) {
const classPrediction = classLabels[i] + ": " + result.scores[i].toFixed(2);
labelContainer.childNodes[i].innerHTML = classPrediction;
}
}, {
includeSpectrogram: true, // in case listen should return result.spectrogram
probabilityThreshold: 0.75,
invokeCallbackOnNoiseAndUnknown: true,
overlapFactor: 0.50 // probably want between 0.5 and 0.75. More info in README
});
// Stop the recognition in 5 seconds. // Load the model
// setTimeout(() => recognizer.stopListening(), 5000); const model = await loadModel();
}
// Classify the audio
const classificationResult = await classifyAudio(model, audioData);
// Display the classification result
const resultSection = document.getElementById("result");
resultSection.textContent = `Class: ${classificationResult}`;
};
// Read the uploaded audio file
fileReader.readAsArrayBuffer(file);
});