HyperSpeech: Text-to-speech addon for HyperSpin 1.x 0.9

[BoOgieman]

Good work guys.

Glad it is solved.

Sent from my Galaxy Note

[potts43]

Good work guys.

Glad it is solved.

Sent from my Galaxy Note

Not sure why UAC causes such a problem but it's strange that it works in win 7 but not win 8 using the RAA.

This is a fantastic add on. Thank you

[classicradios]

I may have to go back to Win 7 64 bit. I really want HyperSpin to auto start on boot up and it's probably easier to shell Windows 7 anyway.

I have the robot voice(s) working too. Using MorphVOX pro and Virtual Audio Streaming, Not free but worth it to me. Thanks to all for the help. HyperSpeech is incredible! Love it.

Jim

[BoOgieman]

I may have to go back to Win 7 64 bit. I really want HyperSpin to auto start on boot up and it's probably easier to shell Windows 7 anyway.

I have the robot voice(s) working too. Using MorphVOX pro and Virtual Audio Streaming, Not free but worth it to me. Thanks to all for the help. HyperSpeech is incredible! Love it.

Jim

Jim,

Thanks for the compliment, I am really glad you sorted it out and that you like the addon.

Even better is that you've managed to take it to the next level with MorphVox and by using audio routing software.

I'm really curious to see the result. Maybe you can post a video someday?

Cheers,

Michael

[potts43]

I've done the same...I'll try to post a video over the holidays!

Merry Christmas

[Polemicist]

Jim,

Thanks for the compliment, I am really glad you sorted it out and that you like the addon.

Even better is that you've managed to take it to the next level with MorphVox and by using audio routing software.

I'm really curious to see the result. Maybe you can post a video someday?

Cheers,

Michael

I'm just really happy to see that it is still being worked on. So many other little gems have been dropped over time. Thanks for your hard work Michael and Merry Christmas to you and yours.

[gigapig]

I've done the same...I'll try to post a video over the holidays!

Merry Christmas

I'd love to see that. I love Hyperspeech, it adds a whole new layer to my setup.

I've got a bunch of voices from Natural Reader and loquendo, but something more robotic would be cool.

So to achieve that I would need both pieces of software that have been mentioned, or just morph vox?

Thanks guys and have a jolly Merry Christmas.

[potts43]

I'd love to see that. I love Hyperspeech, it adds a whole new layer to my setup.

I've got a bunch of voices from Natural Reader and loquendo, but something more robotic would be cool.

So to achieve that I would need both pieces of software that have been mentioned, or just morph vox?

Thanks guys and have a jolly Merry Christmas.

Yeah you need both. You select the virtual cable in the tts output and in cable as input in vox. Works really well. I can't get loquendo to work. It won't allow me to select the installed voices.

[gigapig]

Yeah you need both. You select the virtual cable in the tts output and in cable as input in vox. Works really well. I can't get loquendo to work. It won't allow me to select the installed voices.

Thank you Sir.

For Loquendo you would need to pop this in search C:\Windows\SysWOW64\Speech\SpeechUX\sapi.cpl

Click on the sapi.cpl, then you should be able to change them if they are installed.

http://www.nextup.com/phpBB2/viewtopic.php?f=13&t=6197

[BoOgieman]

Yeah you need both. You select the virtual cable in the tts output and in cable as input in vox. Works really well. I can't get loquendo to work. It won't allow me to select the installed voices.

Mate,

You need to use the 32 bit SAPI ocntrol panel to enable the 32 bit voices onyour 64 bit OS.

That's the way to get the loquendo voices to work.

;-)

C:\Windows\SysWOW64\Speech\SpeechUX\sapi.cpl

LOL,

I Overlooked Gigapig's post...

[classicradios]

Jim,

Thanks for the compliment, I am really glad you sorted it out and that you like the addon.

Even better is that you've managed to take it to the next level with MorphVox and by using audio routing software.

I'm really curious to see the result. Maybe you can post a video someday?

Cheers,

Michael

Sorry for the delay,

Here's a video I made demonstrating the various voices etc..

Jim

https://www.youtube.com/watch?v=JsVowNW3eDk

[djvj]

That british chick near the end sounded the most human. Is that program free or part of some other bundle that lets you change the voices?

[connorsdad]

+1 for the female voice, what package is that voice from ?

Sent from my SM-N9005 using Tapatalk

[gigapig]

Sorry for the delay,

Here's a video I made demonstrating the various voices etc..

Jim

Thanks for sharing, nice cab.

Is that quite simple to add, the MorphVox I mean. I did get the software but haven't had a great deal of time to mess with it.

Thanks.

[BoOgieman]

Thanks for posting this video Jim.

So cool to see your cab with HyperSpeech in action.

You even got the voice morpher hooked up like I suggested.

Playing with the voices is so much fun...

I thought I heard British Kate as the voice. This Loquendo voice is the best sounding voice available imho, but has troubles pronouncing some US English words...

If possible can you make a list of words that need correction?

I will try to find the best match and then update the HyperSpeech DAT file on the ftp.

Also the system bios still need some more work.

All input is welcome.

Cheers mate

Sent from my Galaxy Note

[Trnzaddict]

The British female voice...what is that one? I think the machines sound the best with British females....

EDIT:

Ivona...Amy. That's what that sounds like.

Boogie you might wanna Change over from Loquendo to Ivona they sound a lot more human.

Also did you ever find anything about why game bios are spoken when browsing on genre mode?

[classicradios]

The British female voice...what is that one? I think the machines sound the best with British females....

EDIT:

Ivona...Amy. That's what that sounds like.

Boogie you might wanna Change over from Loquendo to Ivona they sound a lot more human.

Also did you ever find anything about why game bios are spoken when browsing on genre mode?

You are correct! It's Ivona British Amy. I'm using the free trial on her voice right now. I really like her natural sounding voice but would have to pay $45 to keep it after the trial ends. She does pronounce MAME correctly, A word the windows voices fail at pronouncing. Emulator(s) seems to be a word all voices struggle with. I should try a few effects on Amy's voice as I want a slightly sci-fi voice for the cab.

Jim

[BoOgieman]

The British female voice...what is that one? I think the machines sound the best with British females....

EDIT:

Ivona...Amy. That's what that sounds like.

Boogie you might wanna Change over from Loquendo to Ivona they sound a lot more human.

Also did you ever find anything about why game bios are spoken when browsing on genre mode?

Hey mate,

Yes I know what's the matter, sadly it is not something that is going to be fixed neat and easily...

The main reason for this is that HyperSpin does not trigger a LEDBlinky event when pressing the Genre button.

There is also no video's being played in the genre screen, which makes is it not possible for HyperSpeech to detect it is in the Genre screen.

If it was, I could easily make the AutoBio speech stop when going to the Genre wheel...

The only way to do this is to intercept the Genre button press and trigger a stop event for the Bio...

so tell me, what do you use for the genre button/key?

[Trnzaddict]

Hey mate,

Yes I know what's the matter, sadly it is not something that is going to be fixed neat and easily...

The main reason for this is that HyperSpin does not trigger a LEDBlinky event when pressing the Genre button.

There is also no video's being played in the genre screen, which makes is it not possible for HyperSpeech to detect it is in the Genre screen.

If it was, I could easily make the AutoBio speech stop when going to the Genre wheel...

The only way to do this is to intercept the Genre button press and trigger a stop event for the Bio...

so tell me, what do you use for the genre button/key?

Lshift

[gooch]

Hi Boogie, I installed the Ivona Amy voice and it sounds even more natural than the Loquendo Dave voice I was using. Only thing I've found that would need updating in the HyperSpeech.DAT file would be the pronunciation of the years. It puts an "apostrophe" in the middle of it. For example, for 1981 it speaks, "Nineteen apostrophe eighty-one." Let me know if this can be fixed easily, thanks!

[Polemicist]

Isn't there a way of ignoring certain things in the speech engine. Sorry not at home for a week so can't look. In the control panel text to speech options have some way of changing things. That apostrophe coming up would get a little annoying. let us know if you find it.

[BoOgieman]

Hi Boogie, I installed the Ivona Amy voice and it sounds even more natural than the Loquendo Dave voice I was using. Only thing I've found that would need updating in the HyperSpeech.DAT file would be the pronunciation of the years. It puts an "apostrophe" in the middle of it. For example, for 1981 it speaks, "Nineteen apostrophe eighty-one." Let me know if this can be fixed easily, thanks!

Well Greg, every speech engine has its own interpretation of input strings.

The hyperspeech.DAT file is optimized for loquendo voices, so for other voices you need to modify some parts of the .DAT file to fix pronunciation of certain words.

The best way to do this is by using the say.exe command (see my ftp folder) in a dos box, to tweak and test alternations of the word you want to be spoken right.

Then you need to make a new or edit an existing entry in the .dat file.

You can try to locate the regex entry for fixing the year pronunciation and fiddle with that.

As far as I can recall there were issues with years being spoken as numbers and Loquendo voices do no speak the apostrophy literally and this was used as a fix.

Sent from my Galaxy Note

[potts43]

Welll Greg, every speech engine has its own interpretation of input strings.

The hyperspeech.DAT file is optimized for loquendo voices, so for other voices you need to modify some parts of the .DAT file to fix pronunciation of certain words.

The best way to do this is by using the say.exe command (see my ftp folder) in a dos box, to tweak and test alternations of the word you want to be spoken right.

Then you need to make a new or edit an existing entry in the .dat file.

You can try to locate the regex entry for fixing the year pronunciation and fiddle with that.

As far as I can recall there were issues with years being spoken as numbers and Loquendo voices do no speak the apostrophy literally and this was used as a fix.

Sent from my Galaxy Note

Which version of Loquendo ? 6 or 7

[BoOgieman]

Which version of Loquendo ? 6 or 7

I think I'm using 7.5.4

Sent from my Galaxy Note

[evh347]

I can't get this work and I need some help. Like a previous poster several pages back, when I launch Hyperspin, LEDBlinky is not found:

Cannot find LEDBlinky.exe in specified path, setting inactive.

1) Following the Hyperspeech install instructions, I renamed "HyperSpeechProxy.exe" to "LEDBlinky.exe"

2) All of the HyperSpeech files are in a HyperSpeech folder within my Hyperspin root directory.

3) My Settings.ini in my \Hyperspin\Settings\ folder has the following lines added under [LEDBlinky]:

Active=true

Path=H:\Program Files\Hyperspin\HyperSpeech\

4) I added the following two lines to my HyperSpeech.ini under [HyperSpeechSettings]:

RealLedBlinkyActive=true

RealLedBlinkyPath=H:\Program Files\LEDBlinky\LEDBlinky\

Every time I launch Hyperspin, LEDBlinky is not found. The log file located in my \Hyperspin\ root folder confirms the error message I listed at the top of this post.

I'm using Windows 7 64-bit, RocketLauncher, Hyperspin 1.3.2.0, and v.9 of Hyperspeech.

As soon as I change the path of LEDBlinky in my Hyperspin Settings.ini back to "H:\Program Files\LEDBlinky\LEDBlinky\", LEDBlinky launches normally with a launch of Hyperspin.

What am I missing here? I mean, the 2/or 3 steps outlined in the HyperSpeech Readme file seemed pretty straight forward to me.