Skip to main content

Inline Voice Tone & Emotion Tags/Options

Feature Request

Hi, I wanted to suggest a feature that could add a lot more flexibility and expressiveness to the TTS system: inline voice tone/emotion tags/options/buttons.

The idea is to let users control the tone of the voice within a single block of text by wrapping sections in simple tags, or adding options/buttons after highlighting a block of text. For example:

[Calm]This section would be read in a lower, calmer tone.[Calm]

[Angry]This line would sound more intense or aggressive.[Angry]

[Breathy]This would use a softer, breathier delivery.[Breathy]

Tags could represent things like tone, emotion, or delivery style (e.g. Calm, Loud, Quiet, Angry, Breathy, Sad, Excited, Whisper, Serious, etc.).

This would be especially useful for:

- Storytelling and audiobooks

- Dialogue-heavy scripts

- Long-form narration where splitting into multiple clips is inconvenient

- Fine-grained control without switching voices or re-rendering everything

These tags could be optional and safely ignored if unsupported, keeping things backward-compatible and simple for users who don’t need/want them.

Also, the [Tags in this form] are just an example. Simple buttons with even just a few options that add some variation and customization to the voices would be great enough.

I think this would be a powerful addition that makes the TTS output feel much more natural and controllable.

Thanks for considering it, and appreciate all the work you’re doing on the app!

Posted by Nova ·
4 comments

Log in to comment and vote

Comments4

  • grackle_dragoon1g@icloud.com

    •

    Feb 12

    They have SSML tagging on other (admittedly way more expensive) platforms, but I would love this feature too. Or even just the capability to italicize or bold, or use [brackets] that contain instructions that the TTS reader will not read aloud.

    However, I am somewhat familiar with IPA notation, and tried some diacritics(didn’t work, I’m using a multilingual voice and it switched language pronunciations on me), but it WILL use the ˥˦˧˨˩ notations for how high or low the word should be said(˥ means high, ˩ means low), and the English-specific notation of using stress marks (they look like apostrophes and commas, but they’re not): ˈˌ, AFTER the syllable you are wanting to stress (ˌ is used after the syllable/phoneme for secondary stress, ˈ is used after the syllable or phoneme for primary stress).

    In IPA, these marks come before the stressed syllable, in English Heritage Dictionary style, they come after, which makes them English-specific, and I have managed to accomplish quite a bit with these two functioning well within the app, it’s just that most people aren’t going to have an IPA keyboard installed to do this.

  • voice-reader

    Team•

    Feb 11

    Hi Nova,

    Thank you so much for taking the time to share such a thoughtful and detailed suggestion. I really appreciate the creativity behind the idea — inline tone and emotion controls like Calm, Angry, Whisper, etc. could definitely add a new level of expressiveness, especially for storytelling and dialogue-heavy content.


    At the moment, TTS Reader is focused on helping people read books easily and affordably, with a simple setup and voices that are “good enough” without requiring complex configuration. To keep the app accessible and straightforward for most users, we try to avoid adding features that would introduce more setup steps or make the experience more complicated.


    That said, I’m very open to discussing this further.


    If you happen to know of another app that offers this kind of inline tone control at a good price, I’d genuinely love to take a look at how they’ve implemented it. Feel free to share — I’m always interested in learning from what others are doing well.


    Also, we’re planning to release a separate app in the future that’s more focused on advanced customization — including features for users who want high levels of control and even the ability to publish and export their work. It sounds like that direction might be a great fit for what you’re describing.

    If you’re interested, let me know in the comment, I’d be happy to add you to the beta list when it’s ready.


    Best regards,
    Steven

    • grackle_dragoon1g@icloud.com

      •

      Feb 12

      I just commented to Nova, and provided my own solutions to this problem for those who it is really important for, a little more effort on their end, but doesn’t place the burden on the app to accommodate more complexity.

    • Nova

      •

      Feb 12

      There is an app called SpeechWave: Expressive AI TTS. When I found it, it was free to use, though they seem to have switched to a monthly subscription-based model of $2.99 for a basic plan and $9.99 for Plus. (Approximately 37 and 125 minutes of audio respectively)

      So, this has 2 modes. There is a relatively simple mode with buttons for emotion & tone. Literally buttons labeled Calm, Serious, Excited, Friendly, etc. There are options for Speaking Style and even accent. The speaking style is where you select if you want the voice to whisper, to sound energetic, to sound conversational etc. That's the simple mode, you choose different options and that's that.

      The second mode is more advanced. You have to add the tags to the text yourself. So, things like [Angry], [Excited], [Sad], [Whisper], [Laugh], [Sigh], etc. It also has some sliders for voice pitch and speed. But all the customization, aside from pitch and reading speed, is done using tags.

      And yes, that sounds awesome, I’m definitely interested! 🙌 Please add me to the beta list — I’d love to check it out when it’s ready. Looking forward to it!

      Here's some screenshots of the app: