Text-to-Speech with Web Speech API

7shi

This is a stand-alone entry of what is written below.

https://forum.mylittlewordland.com/d/3-what-features-would-you-like-to-see/26

I propose to implement Text-to-Speech support with Web Speech API.

Benefits

Reduced workload for the creator
No additional cost

Disadvantages

Environment-dependent

In summary

Existing unvoiced courses imported from Memrise can also be automatically made voice-enabled!
No cloud contract is required due to the web standard, but it is environment-dependent.

For example, Android requires preparation to install additional voices; Edge and Chrome for PC use online voices provided by Microsoft and Google, so no additional configuration is required. This is especially useful with Edge, where a large number of languages are supported.

You can use the script below to confirm what voices are available in your environment.

https://codepen.io/7shi/pen/rNOdqmq

neoncube

Thanks for splitting this out into a separate topic! This looks awesome! ^_^

7shi

Sorry, I wrote the wrong API name.

(wrong) Web Audio API

(correct) Web Speech API

scmelville

Sorry for my ignorance - would this work similarly to the current Crome add-on "Memrise Audio Uploader"? This looks at what is in the first column and then automatically attaches an MP3 from the Google Translate API.

7shi

scmelville No. Web Speech API is a way to play the voice without uploading the audio.

scmelville

7shi Thanks - so would only work with access to internet? Sounds a more slim-line solution than adding mp3s at any rate.

7shi

scmelville It is environment-dependent, as each browser offers a different type of service. Online voice is only via the Internet, but the API can work offline with using the OS function. (I don't use Apple products, so I don't know about the iPhone situation.)

7shi

This is my sample of the Web Speech API. This will give you an idea of how it works.

https://codepen.io/7shi/pen/PoVeOov