To test locally, your development environment must be accessible from the public internet - this allows our servers to fetch and process your page. Use a tool like ngrok or set up a
Cloudflare Tunnel to expose localhost via a secure, public URL.
You can create filters to exclude content that should not be extracted from your web page. An exclude filter will exclude certain elements, whereas an include filter will exclude everything except the specified elements.
It’s more difficult to force our content extractor to include specific content
once it has determined that it probably isn’t part of the article. However, it
sometimes helps to exclude all other content using filters (see above).Otherwise, please contact support@beyondwords.io and we will do our best to
help. If you need precise, fine-grained control then it might be worth
considering other options such as the
RSS or
API integrations.
Images are considered part of your article’s content and will also be extracted.
Images will show in the Editor relative to the text content of your article
and will be included in generated videos.You can use data attributes
to explicitly set the feature image.
Why can I see content in the dashboard but not in the player?
We automatically extract the publish date from your content by checking various
metadata tags on your HTML page. If the publish date is in the future, then
audio won’t be made available until that time in the player.You can add data attributes
to your HTML to control publish date more precisely and set various other
attributes such as the feature image.
This is normal. It means that no audio is available yet and the player has been
hidden. If Magic Embed is enabled then audio generation will be triggered and
load into the player once it’s ready.
How long does it take for audio to load in the player?
Usually five minutes after the first visitor. The content extraction and audio
generation happen more quickly but the player endpoint is cached for five
minutes. We support cache invalidation on our Enterprise plans.
Yes, Magic Embed will periodically extract content from your page and check if
it has changed. This only happens in response to the player being loaded on the
page and is subject to rate limits to avoid lots of requests.
The simplest approach is to only embed the player on articles published after a certain date. You can check the publish date of your article and conditionally include the player embed code only for newer content.Alternatively, you can set clientSideEnabled: false when initializing the player if the publish date of your article is before some cut-off date.
Does Magic Embed work with single-page websites (JavaScript)?
Yes, our content extractor renders your page with full JavaScript support. It
waits a short amount of time for your page to settle and checks that the network
is no longer active before processing the HTML further.This means that it will work with most single-page applications (SPAs) such as
client-side React apps, Vue.js and Angular. If you load content dynamically from
an API then the content should be extracted.