Explore muse.ai features and learn how to use this video hosting platform to its full potential.
Upload videos and start your journey on muse.ai.
Manage your videos, create collections, channels, set visibility controls, and more.
Search for anything in your collection or within a single video.
Share your video with anyone.
Create custom labels and edit automatic detections with ease.
Embed your videos anywhere on the web.
Clean and powerful video player with in-video search.
Embed searchable collections anywhere on the web.
Ready-made HTML pages with embedded videos, transcriptions, descriptions and more.
Access our APIs and build your own solutions.
To get videos on muse.ai, you need to upload them first.
You can do this by dragging-and-dropping any video into the muse.ai grid, or simply click on the upload button on the top-right corner.
Organize and manage all your video using our state-of-the-art tools.
Easily rename, download, change visibility, share and delete videos using the muse.ai interface.
Collections are groups of videos you can create and share with anyone.
Use the Basic Operations.
A channel is a public view of a collection. To create a channel, simply create a collection and make it public.
You can use channels to showcase videos to people who don't have a muse.ai account, or simply to share videos in a stylish way.
Channels are designed for viewing purposes only, and do not include managment controls. To manage videos in a Channel, go to it's corresponding Collection and use the Basic Operations.
Narrow down your search to a certain category or categories by using search shortcuts.
| Speech | s:word | Searches for words that are said in the video |
| Faces | f:word | Searches for people whose face is visible in the video |
| Text | t:word | Searches for text that is visible in the video |
| Objects | o:word | Searches for objects that are visible in the video |
| Actions | a:word | Searches for actions that are performed in the video |
| Sounds | z:word | Searches for sounds that can be heard in the video |
| Title | t:word | Searches for titles of videos |
| Description | d:word | Searches for descriptions of videos |
Edit automatic transcriptions, label using intuitive interfaces, and cut clips to share with anyone.
On muse.ai, you can label anything.
Save and share the best moments of any video by cutting clips of its highlights.
Stream videos in 4K with DASH/HLS streaming, cast to any screen using Chromecast or Airplay, and search for anything inside your videos using the muse.ai HTML5 responsive video player.
[space] / k |
Play / Pause |
l |
Skip 10 seconds |
j |
Back 10 seconds |
. |
Next frame |
, |
Previous frame |
► |
Skip 5 seconds |
◄ |
Back 5 seconds |
▲ |
Volume up |
▼ |
Volume down |
f |
Toggle fullscreen |
s |
Focus search field |
Bring muse.ai technology to your own websites by using our embedded player and search.
Clean yet powerful embedded video player you can customize to your needs.
To embed our player on your website you only need two lines of code:
<div class="muse-video-player" data-video="y8zH87z" data-width="600"></div> <script src="https://muse.ai/static/js/embed-player.min.js"></script>
Simply set the data-video value to your video id, which you
can find in the address bar of your video on muse.ai.
If you have more than one video on your page, you only need to include the
<script> tag once. It's also a good practice to place it
in the <head> section or at the bottom of the
<body> section.
You can configure the player size by setting the data-width or
data-height attributes. If neither of them is set the player
will default to 640x360.
You can configure the following values:
data-video | Video ID. |
data-width | Player width. |
data-height | Player height. |
data-autoplay | Set to "1" to play the video automaticaly. |
data-loop | Set to "1" to restart the video after it finishes. |
data-volume | Set volume to a value between 0 and 100. |
If you want more control over the player and its interactions with your website, you can use the JavaScript library. This allows you to programmatically create the player and access more of its functions.
To use the JavaScript library you will first of all need to include the following embedded player script (see "Simple usage"). Then, creating an instance of a player is easy:
const player = MusePlayer({
container: '#your-container',
video: 'y8zH87z',
width: 400
});
Using the player object you can perform basic operations on
your video. For example, to pause the video, call player.pause().
MusePlayer parameters:
container | HTML element that will contain the player. |
autoplay | Set to true to play the video automatically. |
loop | Set to true ro restart the video after it finishes. |
volume | Set volume to a value between 0 and 100. |
width | Desired video width. |
height | Desired video height. |
showTitle | Set to false to hide the title. |
showSearch | Set to false to hide the search functionality. |
links | Set to false to disable links to muse.ai. |
sizing | Set to 'fill' to
indicate that the player should fill the entire container's size and
introduce black bars as necessary. Set to 'fit' to span
the parent container while maintaining the player's aspect ratio. If
this value is set, width and height will be
ignored. Default: null. |
logo | Set to false to hide branding on the player. |
MusePlayer methods:
play() | Play the video. |
pause() | Pause the video. |
seek(time) | Skip to the specified time. |
on(event, listener) | Set up a function that will be called when the specified event happens. |
MusePlayer events:
metadata | Player metadata (such as title, description, width, and height) is ready. The metadata object will be passed as the first argument to the listener. |
With muse.ai embedded video search, you can bring our state-of-the-art video search to any website. Allow your users to navigate your video collection, easily find their topics of interest and get insights that would otherwise take hours or days to discover.
A picture is worth a thousand words, so why not check out some of our demos first:
Conference videos University lectures
You can add video search capabilities using our JavaScript library.
First, load the library:
<script src="https://muse.ai/static/js/embed-player.min.js"></script>
Then, initialize a MuseSearch object by telling it about your website layout.
<script>
MuseSearch({
collection: 'TmB8Lt7',
containerInput: '#your-input-field',
containerResults: '#your-result-box'
});
</script>
You can configure the following options:
collection*file
Collection ID of the video collection you want to embed. You can find it in the URL of a collection.
containerInput*string
CSS selector for an input field where users will type their queries.
containerResults*string
CSS selector for an element that will contain search results.
containerPlayerstring
Optional CSS selector where a video will appear after user clicks one of the search results. If this value is not provided, clicking a result will display the video in a lightbox.
defaultInputStylebool
Set this to false to not apply the default search field styling. You will need to style it yourself.
defaultResultsStylebool
Set this to false to not apply the default search result styling. You will need to style it yourself (this might involve some work).
* required
As an alternative to embedding videos on your website, you can simply download a ready-made fully responsive HTML file and use it on your own website as a standalone page.
Simple page that has the video centered both vertically and horizontally.
Example<body>
<div class="container">
<div id="video-player"></div>
</div>
</body>
<script src="https://muse.ai/static/js/embed-player.min.js"></script>
<script>
const player = MusePlayer({
container: document.getElementById('video-player'),
video: 'c41FczZ',
autoplay: true,
loop: true,
volume: 0,
sizing: 'fit',
});
</script>
Simple page that has the video, the video title, and video description, horizontally centered.
Example
<body>
<div class="container">
<h1 id="video-title"></h1>
<div id="video-player"></div>
<h2 id="video-description"></h2>
</div>
</body>
<script src="https://muse.ai/static/js/embed-player.min.js"></script>
<script>
const player = MusePlayer({
container: '#video-player',
video: 'c41FczZ',
autoplay: true,
loop: true,
volume: 0,
width: 'parent',
});
player.on('metadata', metadata => {
document.getElementById('video-title').innerText = metadata.title
document.getElementById('video-description').innerText = metadata.description
});
</script>
Simple two-pane layout with the video and its interactive auto-generated transcipt.
Example
<body>
<div class="container">
<div id="video-player"></div>
</div>
<div class="container-transcript">
<div class="header-transcript">
<h2>Transcript</h2>
</div>
<div id="transcript-list">
<table><tbody id="speech-container"></tbody></table>
</div>
</div>
</body>
<script src="https://muse.ai/static/js/embed-player.min.js"></script>
<script>
const player = MusePlayer({
container: '#video-player',
video: 'VBdrD8v',
autoplay: true,
width: 'parent',
});
// Highlight the current word.
player.on('timeupdate', function(time) {
document.querySelectorAll('.word').forEach(function(word) {
if (Number.parseFloat(word.dataset.start) <= time && Number.parseFloat(word.dataset.end) > time) {
word.classList.add('current');
} else if (word.classList.contains('current')) {
word.classList.remove('current');
}
});
});
fetch(
'https://muse.ai/api/files/i/speech/VBdrD8v', {mode: 'cors'}
)
.then(function(response) {
return response.json();
})
.then(function(words) {
// Sort words by time.
words.sort((a, b) => a[0] - b[0]);
// Group into phrases.
let phrases = [];
let wordPrev;
words.forEach(word => {
let paragraphStart = word[2] == '\n' || wordPrev && word[0] - wordPrev[1] > 1.2;
if (phrases.length == 0 || paragraphStart && wordPrev[2].trim()) {
phrases.push([]);
}
if (word[2].trim()) {
phrases[phrases.length - 1].push(word);
}
wordPrev = word;
});
// Draw phrases.
let list = document.getElementById('speech-container');
phrases.forEach(phrase => {
if (!phrase.length) {
return;
}
let time = phrase[0][0];
let timeLen = time > 3600 ? 8 : (time > 600 ? 5 : 4);
let timeStart = new Date(time * 1000).toISOString().substr(19 - timeLen, timeLen);
let wordsHTML = '';
phrase.forEach(word => {
wordsHTML += `<span class="word" data-start="${word[0]}" data-end="${word[1]}">${word[2]}</span> `;
});
list.insertAdjacentHTML('beforeend', `
<tr>
<td class="phrase-time">${timeStart}</td>
<td class="phrase">${wordsHTML}</td>
</tr>
`);
});
});
// Seek video to word time.
document.addEventListener('click', e => {
if (e.target.classList.contains('word')) {
player.seek(Number.parseFloat(e.target.dataset.start));
}
});
</script>
Simple two-pane layout designed for users who want to provide clickable chapters for navigating a video.
Example
<body>
<div class="container">
<div id="video-player"></div>
</div>
<div class="container-chapters">
<div class="header-chapters">
<h2>Moments</h2>
</div>
<div id="container-list">
<ol>
<li data-start="0.0">Moment One (0s)</li>
<li data-start="2.5">Moment Two (2.5s)</li>
<li data-start="9.7">Moment three (9.7s)</li>
</ol>
</div>
</div>
</body>
<script src="https://muse.ai/static/js/embed-player.min.js"></script>
<script>
const player = MusePlayer({
container: '#video-player',
video: 'c41FczZ',
autoplay: true,
loop: true,
volume: 0,
width: 'parent',
});
document.querySelectorAll('#container-list li').forEach(li => {
li.addEventListener('click', e => {
player.seek(Number.parseFloat(e.currentTarget.dataset.start));
});
});
</script>