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> | |
| <style> | |
| body { | |
| font-family: Arial, sans-serif; | |
| background-color: #f4f4f4; | |
| padding: 20px; | |
| } | |
| .container { | |
| max-width: 700px; | |
| margin: 0 auto; | |
| background: #fff; | |
| padding: 20px; | |
| box-shadow: 0 0 10px rgba(0, 0, 0, 0.1); | |
| } | |
| h1 { | |
| text-align: center; | |
| } | |
| .button { | |
| background-color: #e8b62c; | |
| color: white; | |
| padding: 10px 20px; | |
| text-align: center; | |
| cursor: pointer; | |
| border: none; | |
| margin-top: 10px; | |
| display: block; | |
| width: 100%; | |
| } | |
| .audio-upload { | |
| margin-top: 20px; | |
| text-align: center; | |
| } | |
| .result { | |
| margin-top: 20px; | |
| } | |
| </style> | |
| </head> | |
| <body> | |
| <div class="container"> | |
| <h1>Audio Transcription and Similarity Checker</h1> | |
| <div id="original-audio" class="audio-upload"> | |
| <h2>Upload Original Audio</h2> | |
| <input type="file" id="originalFile" accept="audio/*"> | |
| </div> | |
| <div id="user-audio" class="audio-upload"> | |
| <h2>Upload User Audio</h2> | |
| <input type="file" id="userFile" accept="audio/*"> | |
| </div> | |
| <button id="transcribeButton" class="button">Perform Transcription and Testing</button> | |
| <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; | |
| // Load model and processor | |
| async function loadModel() { | |
| processor = await transformers.AutoProcessor.from_pretrained(MODEL_ID); | |
| model = await transformers.Wav2Vec2ForCTC.from_pretrained(MODEL_ID); | |
| } | |
| async function transcribe(audioFile) { | |
| const arrayBuffer = await audioFile.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 () => { | |
| const originalFile = document.getElementById("originalFile").files[0]; | |
| const userFile = document.getElementById("userFile").files[0]; | |
| if (originalFile && userFile) { | |
| const transcriptionOriginal = await transcribe(originalFile); | |
| const transcriptionUser = await transcribe(userFile); | |
| 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> | |
| `; | |
| } else { | |
| alert("Please upload both audio files."); | |
| } | |
| }); | |
| loadModel(); | |
| </script> | |
| </body> | |
| </html> | |