New Text to Speech experiment - #888

Open

dkotter wants to merge 21 commits into

WordPress:developfrom

dkotter:feature/text-to-speech

Commits on Jul 16, 2026

Add a few new helper methods for TTS
dkotter
committed
178b184
Add a content chunker class used to chunk content down before we do audio processing so we stay under API limits
dkotter
committed
292149e
Add an audio combiner class that takes in multiple audio files and combines those into a single file
dkotter
committed
df7a1f2
Add a speech generator class that takes in some content and sends that to the AI Client to generate speech. Set up to be used for both Ability calls and background jobs
dkotter
committed
73e466d
Add a job manager class that is responsible for the bulk of the work. Fires individual jobs, via cron, that will chunk content down and turn those into audio files, combining all files at the end
dkotter
committed
3523125
Register the TTS experiment and all needed hooks
dkotter
committed
b34c528
Add a TTS REST controller to handle starting jobs and checking on the status of a job
dkotter
committed
b38eddf
Add a generate-speech ability that can be used to generate speech for a string of text or post content from a specific post ID
dkotter
committed
ad66bc0
Add a import-base64-audio ability that takes in base64 encoded data and imports it into the media library as an MP3 file
dkotter
committed
73977b7
Add additional test coverage
dkotter
committed
a358772
Add additional test coverage
dkotter
committed
0c63fd4
Add the necessary client-side code to render both in the editor and the front-end
dkotter
committed
cf878eb
Add E2E tests
dkotter
committed
6c8400a
Add docs
dkotter
committed
38ca34a
Show a loading state immediately after clicking the button
dkotter
committed
db236e6

Navigation Menu

Uh oh!

New Text to Speech experiment - #888

New Text to Speech experiment#888
dkotter wants to merge 21 commits into
WordPress:developfrom
dkotter:feature/text-to-speech

Commits on Jul 16, 2026

Add a few new helper methods for TTS

Add a content chunker class used to chunk content down before we do audio processing so we stay under API limits

Add an audio combiner class that takes in multiple audio files and combines those into a single file

Add a speech generator class that takes in some content and sends that to the AI Client to generate speech. Set up to be used for both Ability calls and background jobs

Add a job manager class that is responsible for the bulk of the work. Fires individual jobs, via cron, that will chunk content down and turn those into audio files, combining all files at the end

Register the TTS experiment and all needed hooks

Add a TTS REST controller to handle starting jobs and checking on the status of a job

Add a generate-speech ability that can be used to generate speech for a string of text or post content from a specific post ID

Add a import-base64-audio ability that takes in base64 encoded data and imports it into the media library as an MP3 file

Add additional test coverage

Add additional test coverage

Add the necessary client-side code to render both in the editor and the front-end

Add E2E tests

Add docs

Show a loading state immediately after clicking the button

Commits on Jul 21, 2026

Merge branch 'develop' into feature/text-to-speech

Ensure we add the title to the post content. Adjust some styling in the admin to make things look better. Add a better loading state and an icon to the button

Better handling for posts that don't have a title set

Don't force MP3 format when we make our request

Add the ability to delete a generated audio file and all associated meta at the post level

Ensure we use the right capability so setting a custom provider works