|
|
<!DOCTYPE html> |
|
|
<html lang="en"> |
|
|
<head> |
|
|
<meta charset="UTF-8"> |
|
|
<meta name="viewport" content="width=device-width, initial-scale=1.0"> |
|
|
<title>Speaker Log Dataset | Open Speech Data</title> |
|
|
<script src="https://cdn.tailwindcss.com"></script> |
|
|
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.0/css/all.min.css"> |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<style> |
|
|
:where([class^="ri-"])::before { content: "\f3c2"; } |
|
|
body { |
|
|
font-family: 'Inter', sans-serif; |
|
|
background-color: #f9fafb; |
|
|
} |
|
|
.hero-bg { |
|
|
background-image: url('https://readdy.ai/api/search-image?query=A%20modern%20visualization%20of%20voice%20data%20analysis%2C%20showing%20sound%20waves%20and%20patterns%20in%20a%20clean%2C%20abstract%20style.%20The%20image%20has%20a%20gradient%20from%20white%20to%20light%20blue%20on%20the%20left%20side%2C%20transitioning%20to%20a%20visualization%20of%20audio%20waveforms%20and%20speaker%20identification%20patterns%20on%20the%20right.%20The%20visualization%20uses%20soft%20blue%20tones%20with%20data%20points%20and%20connections.%20The%20left%20side%20is%20completely%20white%20for%20text%20overlay.&width=1200&height=500&seq=1234&orientation=landscape'); |
|
|
background-size: cover; |
|
|
background-position: center; |
|
|
} |
|
|
.feature-card:hover { |
|
|
transform: translateY(-5px); |
|
|
box-shadow: 0 10px 25px -5px rgba(0, 0, 0, 0.1), 0 10px 10px -5px rgba(0, 0, 0, 0.04); |
|
|
} |
|
|
.download-card:hover { |
|
|
transform: translateY(-3px); |
|
|
box-shadow: 0 10px 15px -3px rgba(0, 0, 0, 0.1), 0 4px 6px -2px rgba(0, 0, 0, 0.05); |
|
|
} |
|
|
</style> |
|
|
</head> |
|
|
<body class="font-sans bg-gray-50"> |
|
|
|
|
|
<nav class="bg-white shadow-sm"> |
|
|
<div class="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8"> |
|
|
<div class="flex justify-between h-16"> |
|
|
<div class="flex items-center"> |
|
|
<div class="flex-shrink-0 flex items-center"> |
|
|
<i class="fas fa-microphone-alt text-indigo-600 text-2xl mr-2"></i> |
|
|
<span class="text-xl font-bold text-gray-900">M3SD</span> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
</nav> |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
<section class="hero-bg relative py-24"> |
|
|
<div class="hero-gradient text-black py-16"> |
|
|
<div class="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8"> |
|
|
<div class="text-center"> |
|
|
<h1 class="text-4xl md:text-5xl font-bold mb-6"> |
|
|
<span class="whitespace-nowrap">M3SD: Multi-modal, Multi-scenario and Multi-language</span> Speaker Diarization Dataset |
|
|
</h1> |
|
|
<p class="text-xl md:text-2xl max-w-3xl mx-auto"> |
|
|
<span class="whitespace-nowrap">A large-scale speaker diarization dataset based on pseudo-labeling</span> |
|
|
</p> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
</section> |
|
|
|
|
|
|
|
|
<div class="max-w-7xl mx-auto px-4 sm:px-6 lg:px-8 py-12"> |
|
|
|
|
|
<div class="bg-white rounded-xl shadow-md overflow-hidden mb-12"> |
|
|
<div class="p-8"> |
|
|
<div class="flex items-center mb-6"> |
|
|
<div class="flex-shrink-0 bg-indigo-100 p-3 rounded-lg"> |
|
|
<i class="fas fa-database text-indigo-600 text-2xl"></i> |
|
|
</div> |
|
|
<h2 class="ml-4 text-2xl font-bold text-gray-900">Dataset Overview</h2> |
|
|
</div> |
|
|
<p class="text-gray-600 mb-6 text-justify leading-relaxed"> |
|
|
Speaker diarization aims to solve the problem of "who speaks when". Existing data resources are often concentrated in specific scenarios such as meetings, resulting in insufficient generalization of the speaker diarization model. The M3SD dataset is a carefully organized speaker diarization dataset with detailed metadata, which aims to promote multi-modal, multi-scenario, and multi-language speaker diarization task research. The dataset contains 770+ hours of conversations, covering multiple scenarios such as online and offline meetings, home communications, outdoor conversations, interviews, movie clips, news broadcasts, and multiple languages including English and Chinese. The data comes from YouTube and is pseudo-labeled through a variety of speaker diarization systems. We will provide audio files, annotation files, and video metadata including uid. |
|
|
You can also download videos from YouTube based on video meta information for multimodal research. The code for data collection has been open sourced: |
|
|
<a href="https://github.com/slwu0209/M3SD" |
|
|
class="text-blue-500 hover:underline" |
|
|
target="_blank" |
|
|
rel="noopener noreferrer"> |
|
|
https://github.com/slwu0209/M3SD |
|
|
</a>. |
|
|
</p> |
|
|
<div class="bg-indigo-50 border-l-4 border-indigo-400 p-4 mb-6"> |
|
|
<div class="flex"> |
|
|
<div class="flex-shrink-0"> |
|
|
<i class="fas fa-info-circle text-indigo-600"></i> |
|
|
</div> |
|
|
<div class="ml-3"> |
|
|
<p class="text-sm text-indigo-700"> |
|
|
This dataset is freely available for academic and non-commercial research purposes. |
|
|
</p> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
|
|
|
|
|
|
<div class="mb-12"> |
|
|
<h2 class="text-2xl font-bold text-gray-900 mb-8 text-center">Key Features</h2> |
|
|
<div class="grid grid-cols-1 md:grid-cols-3 gap-8"> |
|
|
<div class="feature-card bg-white rounded-xl shadow-md overflow-hidden p-6 transition-all duration-300"> |
|
|
<div class="flex items-center mb-4"> |
|
|
<div class="flex-shrink-0"> |
|
|
<i class="fas fa-shapes text-3xl feature-icon"></i> |
|
|
</div> |
|
|
<h3 class="ml-3 font-semibold text-gray-900">Diverse Data</h3> |
|
|
</div> |
|
|
<p class="text-gray-600 "> |
|
|
1,372 records, 770+ hours of data, a large number of different speakers, covering multiple real-world scenarios and multiple languages, and supporting audio-visual multimodal research. |
|
|
</p> |
|
|
</div> |
|
|
<div class="feature-card bg-white rounded-xl shadow-md overflow-hidden p-6 transition-all duration-300"> |
|
|
<div class="flex items-center mb-4"> |
|
|
<div class="flex-shrink-0"> |
|
|
<i class="fas fa-tags text-3xl feature-icon"></i> |
|
|
</div> |
|
|
<h3 class="ml-3 font-semibold text-gray-900">Rich Metadata</h3> |
|
|
</div> |
|
|
<p class="text-gray-600 "> |
|
|
Each recording contains rttm annotations, duration, title, description, video uid and other information to help researchers download and use it. |
|
|
</p> |
|
|
</div> |
|
|
<div class="feature-card bg-white rounded-xl shadow-md overflow-hidden p-6 transition-all duration-300"> |
|
|
<div class="flex items-center mb-4"> |
|
|
<div class="flex-shrink-0"> |
|
|
<i class="fas fa-lightbulb text-3xl feature-icon"></i> |
|
|
</div> |
|
|
<h3 class="ml-3 font-semibold text-gray-900">Novel Process</h3> |
|
|
</div> |
|
|
<p class="text-gray-600 "> |
|
|
Strict data collection and cleaning process, carefully designed data processing process and speaker diarization pseudo-label generation combined with audio and video ensure the reliability of annotation. Relevant code has been open sourced. |
|
|
</p> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
|
|
|
|
|
|
<div class="bg-white rounded-xl shadow-md overflow-hidden"> |
|
|
<div class="p-8"> |
|
|
<div class="flex items-center mb-6"> |
|
|
<div class="flex-shrink-0 bg-indigo-100 p-3 rounded-lg"> |
|
|
<i class="fas fa-download text-indigo-600 text-2xl"></i> |
|
|
</div> |
|
|
<h2 class="ml-4 text-2xl font-bold text-gray-900">Download the Dataset</h2> |
|
|
</div> |
|
|
<p class="text-gray-600 mb-6"> |
|
|
Get immediate access to the M3SD dataset by clicking the button below. The download includes: |
|
|
</p> |
|
|
<ul class="list-disc pl-5 text-gray-600 mb-8 space-y-2"> |
|
|
<li>High-quality audio files in WAV format (16kHz)</li> |
|
|
<li>Speaker diarization pseudo-labels in RTTM format</li> |
|
|
<li>Excel file containing metadata including video uid</li> |
|
|
</ul> |
|
|
<div class="text-center"> |
|
|
<a href="https://pan.baidu.com/s/18xPceIerw2Vbz-ZSvXyDJA?pwd=bgcp" target="_blank" class="download-btn inline-flex items-center px-8 py-4 border border-transparent text-lg font-medium rounded-md text-white bg-indigo-600 hover:bg-indigo-700 focus:outline-none focus:ring-2 focus:ring-offset-2 focus:ring-indigo-500"> |
|
|
<i class="fas fa-download mr-3"></i> Download via Baidu Netdisk (68.65GB) |
|
|
</a> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
|
|
|
|
|
|
<div class="mt-12 bg-white rounded-xl shadow-md overflow-hidden"> |
|
|
<div class="p-8"> |
|
|
<h2 class="text-2xl font-bold text-gray-900 mb-6">Potential Applications</h2> |
|
|
<div class="grid grid-cols-1 md:grid-cols-2 gap-6"> |
|
|
<div class="border-l-4 border-indigo-200 pl-4 py-2"> |
|
|
<h3 class="font-semibold text-gray-800">Speaker diarization model pre-training</h3> |
|
|
<p class="text-gray-600 mt-1"> |
|
|
Add this dataset to train speaker diarization model with better generalization performance. |
|
|
</p> |
|
|
</div> |
|
|
<div class="border-l-4 border-indigo-200 pl-4 py-2"> |
|
|
<h3 class="font-semibold text-gray-800">Audio-visual speaker diarization research</h3> |
|
|
<p class="text-gray-600 mt-1"> |
|
|
Research on multimodal speaker diarization using audio and video. |
|
|
</p> |
|
|
</div> |
|
|
<div class="border-l-4 border-indigo-200 pl-4 py-2"> |
|
|
<h3 class="font-semibold text-gray-800">Further research on speech recognition</h3> |
|
|
<p class="text-gray-600 mt-1"> |
|
|
Transcribe the speech to research automatic speech recognition (ASR). |
|
|
</p> |
|
|
</div> |
|
|
<div class="border-l-4 border-indigo-200 pl-4 py-2"> |
|
|
<h3 class="font-semibold text-gray-800">Explore better semi-supervised methods</h3> |
|
|
<p class="text-gray-600 mt-1"> |
|
|
Research on better semi-supervised methods to generate more accurate pseudo labels. |
|
|
</p> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
|
|
|
<div class="mt-12 bg-white rounded-xl shadow-md overflow-hidden pb-8"> |
|
|
<div class="p-8"> |
|
|
<h2 class="text-2xl font-bold text-gray-900 mb-6">Citation</h2> |
|
|
<p class="text-gray-600 mb-4"> |
|
|
If you find this dataset or code useful in your research, please consider citing the following paper: |
|
|
</p> |
|
|
<div class="bg-gray-50 p-4 rounded-md border-l-4 border-indigo-200"> |
|
|
<pre class="text-sm text-gray-700 whitespace-pre-wrap">@article{wu2025m3sd, |
|
|
title={M3SD: Multi-modal, Multi-scenario and Multi-language Speaker Diarization Dataset}, |
|
|
author={Shilong Wu and Hang Chen and Jun Du}, |
|
|
journal={arXiv preprint arXiv:2506.14427}, |
|
|
year={2025} |
|
|
}</pre> |
|
|
</div> |
|
|
</div> |
|
|
</div> |
|
|
|
|
|
|
|
|
|
|
|
<script> |
|
|
|
|
|
document.querySelector('.download-btn').addEventListener('mouseenter', function() { |
|
|
this.querySelector('i').classList.add('animate-bounce'); |
|
|
}); |
|
|
|
|
|
document.querySelector('.download-btn').addEventListener('mouseleave', function() { |
|
|
this.querySelector('i').classList.remove('animate-bounce'); |
|
|
}); |
|
|
|
|
|
|
|
|
function copyPassword() { |
|
|
navigator.clipboard.writeText('vz5k').then(function() { |
|
|
const copyButton = document.querySelector('button[onclick="copyPassword()"]'); |
|
|
const originalIcon = copyButton.innerHTML; |
|
|
copyButton.innerHTML = '<i class="fas fa-check"></i>'; |
|
|
copyButton.classList.remove('text-gray-500'); |
|
|
copyButton.classList.add('text-green-500'); |
|
|
|
|
|
setTimeout(function() { |
|
|
copyButton.innerHTML = originalIcon; |
|
|
copyButton.classList.remove('text-green-500'); |
|
|
copyButton.classList.add('text-gray-500'); |
|
|
}, 2000); |
|
|
}); |
|
|
} |
|
|
</script> |
|
|
<p style="border-radius: 8px; text-align: center; font-size: 12px; color: #fff; margin-top: 16px;position: fixed; left: 8px; bottom: 8px; z-index: 10; background: rgba(0, 0, 0, 0.8); padding: 4px 8px;">Made with <img src="https://enzostvs-deepsite.hf.space/logo.svg" alt="DeepSite Logo" style="width: 16px; height: 16px; vertical-align: middle;display:inline-block;margin-right:3px;filter:brightness(0) invert(1);"><a href="https://enzostvs-deepsite.hf.space" style="color: #fff;text-decoration: underline;" target="_blank" >DeepSite</a> - 🧬 <a href="https://enzostvs-deepsite.hf.space?remix=OldDragon/m3sd" style="color: #fff;text-decoration: underline;" target="_blank" >Remix</a></p></body> |
|
|
</html> |
