Does your feature request involve difficulty for you to complete a task? Please describe.

When using the TTS feature in the ebook reader for japanese text, the system would read out both the kanji and furigana (ruby annotations), leading to redundant pronunciation.

Describe the solution you'd like

I would like the TTS system to read out the kanji only and exclude the furigana reading. I have implemented a fix in my own local project files. I did this by parsing the book for ruby elements and then build a mapping of rb (base text) and rt (reading text). Then use the mapping to strip out furigana readings from the ssml output.

Describe alternatives you've considered

I tried splitting the range into multiple ranges to exclude rt tags, but it didnt work. Therefore, I went with the solution I mentioned above.

I am not sure if my method is the best, so I would appreciate any advice or input. Thanks!