classifai
classifai copied to clipboard
10up
Integrate Azure Text to Speech for "read-to-me" functionality on blog posts
Is your enhancement related to a problem? Please describe. Azure's Text to Speech is able to convert text to audio in near real time, play it back, and save it as a file for later use. This could be leveraged to assist with accessibility improvements for text-only/text-heavy content so that it could be available as audio content.
Describe the solution you'd like If this concept gains positive feedback, then we'll want to iterate on it to build out more specific requirements before we dive into development on this.
Designs n/a
Describe alternatives you've considered This functionality exists via AWS Labs and WP Engine's Amazon Polly plugin, but would mimic that utilizing Azure instead of AWS.
Additional context none
Latest update from Azure adds 5 additional languages for a total of 54 languages/locales with 73 neural voices and 77 standard voices available.
Latest update from Azure brings totals to 129 neural voices and 77 standard voices across 54 languages/locales plus the ability to create a custom neural voice (currently still needs Azure approval even though this is in GA).
For what its worth, Watson also has a Text to Speech function that could be a good option to compare to Azure to see which might be the "best" service to build into ClassifAI for this. Same with Amazon Polly as an option here.
I think going with Watson is the smart choice since we integrate with it already. They both offer different languages and the speed between their translations seems negligible at best.
@pixeldevsio seems reasonable to me, are you able to work on that as a PR or should I look to someone else to help step in there?
I can work on it. I have some time in the coming weeks.
On Wed, Nov 9, 2022, 7:00 PM Jeffrey Paul @.***> wrote:
@pixeldevsio https://github.com/pixeldevsio seems reasonable to me, are you able to work on that as a PR or should I look to someone else to help step in there?
— Reply to this email directly, view it on GitHub https://github.com/10up/classifai/issues/115#issuecomment-1309561027, or unsubscribe https://github.com/notifications/unsubscribe-auth/AOFN3PWCQIBNS4YIOKCLF4LWHQ3I5ANCNFSM4IIBZ63A . You are receiving this because you were mentioned.Message ID: @.***>
so far we’re generating an mp3 file from the Watson API. TODO:
- Hook up an individual post option to turn the audio on (default to off)
- Hook into the save_post and generate audio files when needed
- Come up with a storage/naming convention for the audio files
- Create a block to display the audio file
- Miscellaneous fixes and testing