Spaces:
Running
Running
| <html lang="en"> | |
| <head> | |
| <meta charset="UTF-8"> | |
| <meta name="viewport" content="width=device-width, initial-scale=1.0"> | |
| <title>Audio Transcription and Similarity Checker</title> | |
| <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/5.15.3/css/all.min.css"> | |
| <style> | |
| body { | |
| font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif; | |
| background-color: #f4f4f4; | |
| padding: 20px; | |
| } | |
| .container { | |
| max-width: 800px; | |
| margin: 0 auto; | |
| background: #fff; | |
| padding: 30px; | |
| border-radius: 10px; | |
| box-shadow: 0 4px 8px rgba(0, 0, 0, 0.1); | |
| } | |
| h1 { | |
| text-align: center; | |
| margin-bottom: 30px; | |
| } | |
| .audio-section { | |
| text-align: center; | |
| margin-bottom: 20px; | |
| } | |
| input[type="file"] { | |
| display: none; | |
| } | |
| .upload-btn { | |
| background-color: #007bff; | |
| color: white; | |
| padding: 10px 20px; | |
| cursor: pointer; | |
| border-radius: 5px; | |
| margin: 10px; | |
| border: none; | |
| display: inline-block; | |
| } | |
| .button { | |
| background-color: #28a745; | |
| color: white; | |
| padding: 10px 20px; | |
| cursor: pointer; | |
| border: none; | |
| border-radius: 5px; | |
| margin-top: 20px; | |
| display: block; | |
| width: 100%; | |
| font-size: 16px; | |
| } | |
| .button:hover { | |
| background-color: #218838; | |
| } | |
| #progress-bar { | |
| width: 0; | |
| height: 20px; | |
| background-color: #4caf50; | |
| text-align: center; | |
| line-height: 20px; | |
| color: white; | |
| border-radius: 5px; | |
| display: none; | |
| } | |
| #progress-container { | |
| width: 100%; | |
| background-color: #ddd; | |
| border-radius: 5px; | |
| margin-top: 20px; | |
| } | |
| .result { | |
| margin-top: 20px; | |
| } | |
| .recorder { | |
| cursor: pointer; | |
| background-color: #dc3545; | |
| color: white; | |
| padding: 10px 20px; | |
| border-radius: 50%; | |
| font-size: 24px; | |
| display: inline-block; | |
| margin-top: 20px; | |
| } | |
| .stop-button { | |
| background-color: #ff5733; | |
| color: white; | |
| padding: 10px 20px; | |
| cursor: pointer; | |
| border-radius: 5px; | |
| display: none; | |
| margin: 10px; | |
| } | |
| </style> | |
| </head> | |
| <body> | |
| <div class="container"> | |
| <h1>Audio Transcription and Similarity Checker</h1> | |
| <div id="original-audio" class="audio-section"> | |
| <h2>Upload or Record Original Audio</h2> | |
| <label class="upload-btn" for="originalFile">Choose Audio File</label> | |
| <input type="file" id="originalFile" accept="audio/*"> | |
| <div id="originalRecorder" class="recorder"> | |
| <i class="fas fa-microphone"></i> Record | |
| </div> | |
| <button id="stopOriginalRecording" class="stop-button">Stop Recording</button> | |
| </div> | |
| <div id="user-audio" class="audio-section"> | |
| <h2>Upload or Record User Audio</h2> | |
| <label class="upload-btn" for="userFile">Choose Audio File</label> | |
| <input type="file" id="userFile" accept="audio/*"> | |
| <div id="userRecorder" class="recorder"> | |
| <i class="fas fa-microphone"></i> Record | |
| </div> | |
| <button id="stopUserRecording" class="stop-button">Stop Recording</button> | |
| </div> | |
| <button id="transcribeButton" class="button">Perform Transcription and Testing</button> | |
| <div id="progress-container"> | |
| <div id="progress-bar">0%</div> | |
| </div> | |
| <div id="result" class="result"></div> | |
| </div> | |
| <script src="https://cdn.jsdelivr.net/npm/@huggingface/transformers"></script> | |
| <script> | |
| const MODEL_ID = "facebook/wav2vec2-large-960h"; // Sample model, change if necessary | |
| let processor, model; | |
| let originalAudioBlob = null; | |
| let userAudioBlob = null; | |
| // Load model and processor | |
| async function loadModel() { | |
| processor = await transformers.AutoProcessor.from_pretrained(MODEL_ID); | |
| model = await transformers.Wav2Vec2ForCTC.from_pretrained(MODEL_ID); | |
| } | |
| // Simulate progress bar loading | |
| function updateProgressBar(percentComplete) { | |
| const progressBar = document.getElementById("progress-bar"); | |
| progressBar.style.width = percentComplete + "%"; | |
| progressBar.innerHTML = percentComplete + "%"; | |
| if (percentComplete === 100) { | |
| setTimeout(() => { | |
| progressBar.style.display = "none"; | |
| progressBar.style.width = "0%"; | |
| }, 500); | |
| } else { | |
| progressBar.style.display = "block"; | |
| } | |
| } | |
| async function transcribe(audioBlob) { | |
| const arrayBuffer = await audioBlob.arrayBuffer(); | |
| const audioData = new Float32Array(arrayBuffer); | |
| const inputValues = processor(audioData, {return_tensors: "pt", padding: true}).input_values; | |
| const logits = await model(inputValues).logits; | |
| const predicted_ids = logits.argmax(-1); | |
| const transcription = processor.decode(predicted_ids, {skip_special_tokens: true}); | |
| return transcription; | |
| } | |
| document.getElementById("transcribeButton").addEventListener("click", async () => { | |
| if (!originalAudioBlob && !document.getElementById("originalFile").files[0]) { | |
| alert("Please upload or record original audio."); | |
| return; | |
| } | |
| if (!userAudioBlob && !document.getElementById("userFile").files[0]) { | |
| alert("Please upload or record user audio."); | |
| return; | |
| } | |
| updateProgressBar(0); | |
| let percentComplete = 0; | |
| const progressInterval = setInterval(() => { | |
| percentComplete += 10; | |
| updateProgressBar(percentComplete); | |
| if (percentComplete >= 100) clearInterval(progressInterval); | |
| }, 200); | |
| if (!originalAudioBlob) { | |
| originalAudioBlob = document.getElementById("originalFile").files[0]; | |
| } | |
| if (!userAudioBlob) { | |
| userAudioBlob = document.getElementById("userFile").files[0]; | |
| } | |
| const transcriptionOriginal = await transcribe(originalAudioBlob); | |
| const transcriptionUser = await transcribe(userAudioBlob); | |
| clearInterval(progressInterval); | |
| updateProgressBar(100); | |
| const levenshteinDistance = (a, b) => { | |
| let dp = Array.from({length: a.length + 1}, () => Array(b.length + 1).fill(0)); | |
| for (let i = 0; i <= a.length; i++) dp[i][0] = i; | |
| for (let j = 0; j <= b.length; j++) dp[0][j] = j; | |
| for (let i = 1; i <= a.length; i++) { | |
| for (let j = 1; j <= b.length; j++) { | |
| dp[i][j] = a[i - 1] === b[j - 1] ? dp[i - 1][j - 1] : Math.min(dp[i - 1][j], dp[i][j - 1], dp[i - 1][j - 1]) + 1; | |
| } | |
| } | |
| return dp[a.length][b.length]; | |
| }; | |
| const similarityScore = 1 - levenshteinDistance(transcriptionOriginal, transcriptionUser) / Math.max(transcriptionOriginal.length, transcriptionUser.length); | |
| document.getElementById("result").innerHTML = ` | |
| <h2>Transcription Results</h2> | |
| <p><strong>Original Transcription:</strong> ${transcriptionOriginal}</p> | |
| <p><strong>User Transcription:</strong> ${transcriptionUser}</p> | |
| <p><strong>Levenshtein Similarity Score:</strong> ${similarityScore.toFixed(2)}</p> | |
| `; | |
| }); | |
| // Initialize model | |
| loadModel(); | |
| // Handle voice recording (using browser APIs) | |
| const recordAudio = () => { | |
| return new Promise(async resolve => { | |
| const stream = await navigator.mediaDevices.getUserMedia({ audio: true }); | |
| const mediaRecorder = new MediaRecorder(stream); | |
| const audioChunks = []; | |
| mediaRecorder.addEventListener("dataavailable", event => { | |
| audioChunks.push(event.data); | |
| }); | |
| mediaRecorder.addEventListener("stop", () => { | |
| const audioBlob = new Blob(audioChunks); | |
| resolve(audioBlob); | |
| }); | |
| mediaRecorder.start(); | |
| return mediaRecorder; | |
| }); | |
| }; | |
| let originalRecorderInstance, userRecorderInstance; | |
| document.getElementById("originalRecorder").addEventListener("click", async () => { | |
| originalRecorderInstance = await recordAudio(); | |
| document.getElementById("stopOriginalRecording").style.display = "inline-block"; | |
| document.getElementById("originalRecorder").style.display = "none"; | |
| }); | |
| document.getElementById("stopOriginalRecording").addEventListener("click", () => { | |
| originalRecorderInstance.stop(); | |
| originalRecorderInstance.addEventListener("stop", async () => { | |
| originalAudioBlob = new Blob(originalRecorderInstance); | |
| alert("Original audio recorded!"); | |
| }); | |
| document.getElementById("stopOriginalRecording").style.display = "none"; | |
| document.getElementById("originalRecorder").style.display = "inline-block"; | |
| }); | |
| document.getElementById("userRecorder").addEventListener("click", async () => { | |
| userRecorderInstance = await recordAudio(); | |
| document.getElementById("stopUserRecording").style.display = "inline-block"; | |
| document.getElementById("userRecorder").style.display = "none"; | |
| }); | |
| document.getElementById("stopUserRecording").addEventListener("click", () => { | |
| userRecorderInstance.stop(); | |
| userRecorderInstance.addEventListener("stop", async () => { | |
| userAudioBlob = new Blob(userRecorderInstance); | |
| alert("User audio recorded!"); | |
| }); | |
| document.getElementById("stopUserRecording").style.display = "none"; | |
| document.getElementById("userRecorder").style.display = "inline-block"; | |
| }); | |
| </script> | |
| </body> | |
| </html> | |