Reading Hidden Codes in Epubs and unvoiced words bestween<>

Hi,
 
This is my first time posting here, and i hope that i am seeking for the correct answer in the correct forum, please bare with me as these maybe some very stupid questions.
 
Some times when i try to play epub books (only EPubs) in @voice, it reads the hidden codes behind the text style..
Example: 
Actual Text    With that in mind, I readily agreed to go without hesitation 
How it reads   With that in mind, I readily agreed to font style equals 2 binds slash slash go without hesitation 
 
As a temp solution i copy and past the chapters in .txt and save as plain text for it to only read the visible text and it takes only small effort but it is annoying when it happens when listening while driving.
So i would like the for tts to read only the visible text from the epub as shown on the screen.
 
another small thing: most of the time the tts doesnt read words between <>. i say most of the time because sometimes it reads it if its <two words> or more
Example:
Actual Text    (and the man was going to <Paris> after he finishes the job)
How it reads  (and the man was going to  after he finishes the job)
 
There's probably a very easy solution to these issues that eludes me giving my ignorance, so please be kind :)
 
Thank you in advance

Hi,
 
Thanks for the quick response,
Here is a link to a novel that you will find the mentioned error in first page (Title Page) and also in some areas in the middle of the chapters. 
Link to GDrive

I barely got the file right now. Listened to the title page, all was read correctly. The text on that page is:
 
and this is exactly what I hear. No extra "hidden text" between <> characters are heard. Maybe you have used @Voice "Edit speech" feature and entered some speech replacements incorrectly? Please disable them all and try to listening to the same page again, then enabling one by one you could identify the offending replacements and fix or delete them.
 
Greg

Yes, epub files are HTML or XHTML files embedded inside a ZIP file (with extension renamed to .epub). Where you actually see a < character on the screen, in HTML source code it has to be encoded as:
 
&lt;
 
and > as
 
&gt;
 
If you need to replace them, you would need to replace these codes instead of literal < or > characters. It's best to double-tap a sentence where they occur in your ebook, then go to "Edit speech" screen, add a new replacement or edit and existing one. You will then see the current sentence that you double-tapped at the bottom of screen as "sample text", with all the codes visible. After you type your replacement, you may press the "TEST: ORIGINAL" button there and see how your replacement worked, what will be sent to the speech engine. You may also press the round loudspeaker button there and listen to that sample.

Dear Greg,
I have a similar problem, but it is more serious.
The program always reads all the HTML code that is hidden in the text.
It reads only correct, if I delete all the formation of the text first.
How can I solve this problem?
Thanks in advance.
Rofa

No, I do not mean speech repalacements, but if the text is bold or italic or it is a headline and so on the problem occurs.
I have no problems with pure text.
But now the problem is less serious after I discovered your advice to reyanoo.
I wrote the code [<font] into the speech replacements and it is much better now.
Here is an example:
 
This is the text I would like to be read:
70.218 Frauen und Mädchen in Deutschland sind beschnitten, das geht aus der Dunkelzifferstatistik hervor, die Terre des Femmes – Menschenrechte für die Frau zum Internationalen Mädchentag am 11. Oktober veröffentlichte.

And this is the text the program actually reads:
70.218 Frauen und Mädchen in Deutschland sind beschnitten, das geht aus der <a href="https://www.frauenrechte.de/images/downloads/presse/dunkelzifferstatistik/FGM-C-Dunkelzifferstatistik-2019.pdf"; class="avar_hl_snt" style="color: rgb(0, 0, 0);">Dunkelzifferstatistik hervor, die Terre des Femmes – Menschenrechte für die Frau zum Internationalen Mädchentag am 11. Oktober veröffentlichte.
 
Greetings from Rofa

I disabled all my 21 speech replacements and it looks as if this was the solution.
Thank you very much!
Rofa

Hello! Useful post. I have a more subtle puzzle to solve. I use a lot of regex to clean up text and remove citations... But for .epub articles from some journals, the HTML often found inside inline citations messes up regex matching. Here is an example article and my settings files: https://drive.google.com/...qusbueG8KG1mkZIKw6pD_P