Custom Voice Pack

[AkitoKamimura]

hey, im not sure if this belongs here, or if this question got even asked.
Im a bit unhappy with the ammount of voices there is in the normal version for Voice Attack (probably cause the only voices it gives me are the from windows). But is there a way, to have a custom voice pack selfmade, where i can just type and it will read whatever i just typed, i only have seen those, that just respond with a prerecorded file, to a specific spoken command.

I do need a custom voice, cause i am unable to talk with people online, and text to speech has helped me out a lot, but the voices sound so, robotic and in the game i play, a lot of people use those same voices, wich makes it hard to say who just said something.
If anyone does know how to make a fully working custom voice, i would be interested to find out how they are made, and what i need to also make one for myself.
Greetings, and thanks for any help.

[Pfeil]

"Voice Pack" would generally refer to a third-party HCS Voice Packs product, which uses prerecorded sound files.

What you seem to be looking for are text-to-speech voices, which interpret text to produce sound.


There are a number of third party providers of TTS voices out there, and Exergist has a list on the Discord:

3rd Party Text-to-Speech Voices
NextUp
AT&T Natural Voices = https://nextup.com/attnv.html
Ivona Voices = https://nextup.com/ivona/ (multi-voice packs available)
Harpo Software (free trial and multi-voice packs available)
Ivona Voices = https://harposoftware.com/en/2-main/s-1/index/brand-ivona/category-single_voices
Nuance Voices = https://harposoftware.com/en/2-main/s-1/index/brand-nuance/category-single_voices
CereProc
CereProc Voices = https://www.cereproc.com/en/storesapi (see customizable demo at top of page)

Note that 100% compatibility is not guaranteed since these are 3rd party after all. You assume full responsibility for what happens to your PC if you employ any of these voices. If you encounter issues with any voices we'll update this list to reflect it.

There are also some TTS voices that are technically compatible with SAPI, but don't officially support it:

Unlock More Native Text-To-Speech Voices
https://www.ghacks.net/2018/08/11/unlock-all-windows-10-tts-voices-system-wide-to-get-more-of-them/
You assume full responsibility for what happens to your PC if you employ this method

There are some companies that offer services to create custom text-to-speech voices, but that is likely to be an expensive process.
There are also some (open-source) projects that provide tools for this, however they are fairly complex.
Either of these would require an actual person to speak large amounts of example text.

You could look into having a custom SAPI TTS voice created for you by a third party, but all of this is outside the scope of VoiceAttack itself.

[AkitoKamimura]

Thanks alot allready, this is helping me out a lot.
I did knew that Voice attack wasnt there for the voices, just for the programm itself.
But realy, thanks agian this will probably solve my problems now, since i didnt even knew where to start

[iceblast]

Ivona 2 Voices are very good. You can install them, and either use Voiceattack to read the text using the voice you install, or use a program that's called Balabolka, which is one of the best text to voice programs out there. I've been using it for years. I use it to read books all the time.

My favorite Ivona 2 Voice is Joey. For most people setting the speed to 2 or 3 is best, for I listen at 6 most of the time.

You can easily make commands with VA that will say things for you.

With VA, it's very easy to setup a way to send text to Balabolka and have it read comment sections, websites, and basically any text you copy to clipboard.

Balabolka is free, and can easily turn any text into a mp3.

With VA by itself, it can read text for you, but with Balabolka you have more control, like being able to pause, and unpause as needed.

[AkitoKamimura]

I have heard about balabolka, but wasnt so sure on how to use that.
Also, is there a way to use balabolka and voice attack in the following way.
Convert Speech to text with VA, send the text over to Balabolka, let balabolka read the text.
And that all in one go, without a long delay inbetween?
Would be very cool to know if thats possible, and if, how?
thanks for that extra information

[iceblast]

You want to speak and have it turned into text, then have balabolka read it.

VA does have Dictation, but it's not really that great, but that's the fault of windows, not VA.

If you type out what you want said, and copy it to the clipboard, you can have either VA, or Balabolka read it out loud for you.

Probably one of the best options for speech to text is Dragon Naturally Speaking. I use it all the time, but even then it's hit and miss on how well it hears what your saying, but it making more mistakes then I'm sure you would wish. Voice to Text just isn't quite good enough yet, but it is getting better all the time.

If you don't mind typing, you can do speech that way with np though.

With VA, you can very easily make lots of commands that can be read out for you. It's very simple in VA to do that.

[AkitoKamimura]

Ahh okay.
I have managed to setup some basic stuff for VA, where i have to time when to say something, and then it can translate word for word into tts.
The most annyoing about this is, i have to constantly add new words i want to use, before i am able to use them.
but i did also found out just typing it in is the best solution, even tho it is a bit slow sometimes <.<

But thanks for all the additional Information