Just before the weekend, we released a new version of Pro Editor with a few commonly requested improvements.

What’s new

• Added request headers in the Pro Editor to improve how network calls are handled
• Improved login system by connecting to a new (faster) permissions service
• Library filters are now persisted between sessions
• Added support for expressive voices in the Pro Editor

Fixes & improvements

• Fixed issues with login redirects and authentication loops
• Fixed a bug where the editor could get stuck loading if there was no active session
• Fixed problems with library filters not behaving correctly
• Fixed issues in sound design callbacks
• Fixed small UI issues like copying tags in the library
• Removed unnecessary backend calls to improve performance
• Hid controls that were not actually working to avoid confusion

The endpoint to retrieve an audioform result has been updated.

What has changed:

The version header is documented to be mandatory, and this is now enforced.

How does this work

You need to specify the version number (1, 2, or 3) you wish to get the result in.
Historically, if you didn't specify that version you would get an early alpha version (version 0.0.1).

What are the users required to change:

Early adopters have used this endpoint in beta form, and are working with an early, undocumented, version 0.0.1. They would expect the call to succeed and return a result in this early developement version.

The change means that their calls will now fail with a 422 status code if they aren’t setting the version header in the /audioforms GET request.

They need to set the version header to 0.0.1 and don’t need to change the rest of the code. They are also encourage to migrate to version 3.

Scheduled maintenance will begin on Sunday at 2026-03-15 11:00 UK time and is expected to last 6hrs. This is to support some significant upgrades to the AudioStack ecosystem which will improve loading speed and performance.

During this time:

1. The API will undergo intermittent outages
2. The platform will undergo intermitted outages
3. Users will not be able to create new orgs or users.

🚧

We’re scheduling a planned maintenance window on Saturday, 7 Feb 2026, from 09:00–11:00 GMT.

Duration: up to 2 hours Impact: Temporary disruption to some API services may occur Why this timing: Scheduled outside standard working hours to minimise impact on our clients

We’ll share updates if anything changes and confirm once maintenance is complete on the status page.

UX improvements:

• Changed the copy in Create an Ad to make it clearer what the voice preview does.

Technical improvements:

• Updated AudioForm to use latest version - improving performance.

We also made a number of improvements to products currently in beta.

Please get in touch if you have any feedback.

Today we shipped a performance improvement to our backend which makes our mastering engine faster than ever and reduces the likelihood of failed API calls. This change benefits both API customers and those using the AudioStack platform, with mastering requests being around 75% faster at scale.

In this release, we made an improvement to length management in Create an Ad, to make it easier to fit your content within a fixed length ad while editing your work (more improvements to come on this in the new year!)

We also made a few small bug fixes and improvements across the platform:

• Fixed a crash when clearing your search in the sound effects library in Spanish
• Fixed a loading issue when you have thousands of sessions in a single project
• Added "use" button to rows in the sound library, to make it easier to select a sound template

In this update:

• Fixed a bug where speech could not be unselected in Pro Editor.
• Fixed a bug where the Speech Editor was remaining open in Pro Editor while the ad was generating.
• Improved the UX of voice selection in pro editor so that a more relevant voice is displayed by default.
• Fixed a bug with the media file browser in Pro Editor to ensure the contents are not cut off when listing many items.

We also made a significant improvement to generated ads in English (affecting Create an Ad and Instant Ad, plus API usecases using the brief endpoint, meaning that generated scripts are more likely than ever to be the correct length for the ad.

Today we released:

• An improvement to the language selection control in Create an Ad - to make it easier to choose the correct language for your audio ad.
• A security update to our contact form.
• Removed an integrated survey tool due to bugs with the provider software - to ensure stability unaffected.
• Fixed some broken website links.

In this release, we fixed:

• An issue with downloading WAV files that meant they were not downloading properly from Create an Ad. This required a small change to how you download a WAV - simply select the WAV option from the dropdown in Advanced Settings then hit generate to download your ad in this format going forward.

• We removed a few broken links, including a link to an old service "Audio Evaluation Service" from the File manager page.
• Restored the ability to choose a save location for your recording in the Recording Booth.

• Fixed a crash which occurred when zooming in on the advanced controls, when using Create an Ad in Spanish.
• Fixed an issue where page numbers displayed incorrectly in the Projects page for users with a very large number of sessions in their project.
• Removed a legacy feature (inviting external voice actors using an automated task management system) from the Voice Manager page for admin users.

We also pre-filled the Organization ID for users being invited to the platform, so that they no longer need to fill this in themselves, reducing the risk of unnecessary free trial organisations being created.

Tomorrow, we will be releasing a user experience improvement to make it easier to select your script language - watch this space 👀