LONGECITY

What's a good Text-to-Speech/ Human-speech-synthesizer Software? Are they at the level of The Terminator Movie-quality yet? If you remember from terminator 1,2,3 or Sarah Connor Chronicles, the machines/Terminators can generate different human-voices to fool the real humans (there were even shape-shifting mecury-type terminator-models in T-2 and T3 that can impersonate any person, but that's further in future).

a multitasking tip: when doing work that does not require too much thinking/creativity, listen to stuff. I'm a fan of audiobooks. listen to them and other shows/podcast etc rather than music when working/studying , esp at my computer. I've reached a pt where I can do tedious tasks/problems while also absorbing info through the background speech, and like to fall asleep to books, tho sometime the stuff is so nice i spend a hr lying there listening.

I wonder where're some freely available or good quality,inexpensive Text-to-speech softwares to generate human-quality sounds. I know my Adobe 8 has the Read-a-Loud fxn , and i've used it with some e-books, not as satisfactory, but a nice way to listen to some stuff, but only if they're in the pdf format tho. I'd like one that can read any file, any txt, any web-page, real-human-voice database.

http://www.neospeech.com/ is very good IMO. Its not free unfortunately.

Cepstral makes good microsoft tts plugins that are free to demo and about 30 dollars per voice. The only problem i've found is that they aren't added to windows like the default tts voices are. They're kept separate so they can be activated once you buy a license. The problem is that there is a time delay of a second or two before the voice actually starts to read text for you.

I wouldn't say they're terminator quality either. I don't really believe any voice synth is there yet. There are real subtleties to the human voice. It'll take some really fancy R and D aka money dumped into it or an AI behind it before you truly hear machines that can talk convincingly like people.
Edited by bobscrachy, 01 December 2008 - 08:06 PM.

I think I'm as much of an auditory learner as visual learner, if not more auditory in learning style.

Besides, there's a thing as eye fatigue, damage, irritation, dryness, and learning utilizing all senses would work a bit better

IMO this is the best one on the net with 16 diff voices ... and its free!

http://www.research....eb/tts/demo.php

The synthesis still isn't great, but you may find some that are acceptable to your ears.

I've been using the UK voice on Natural Reader 7. To me the UK accent sounds most "normal", that's why I am using that one even though I'm American... you can test some out on their site: http://www.naturalre...s.com/index.htm

It's easy to copy paste text into it, and export to MP3, which I then export to my MP3 player...I had some manuscripts I listen to while I am working out (and also some articles from websites). Not great quality, but acceptable.

Also with that program, you can add music in the background (it merges it when it encodes the MP3, so the final MP3 can have music and speech), which I do at a low volume. Imo, it helps make it seem less annoying/artificial. I think it costed about $35.

There is a text-to-speech freeware named panopreter, it works very well on Windows Vista, you don't need to buy or install any other voice engine if you are a vista user. It reads with Microsoft Anna voice on Vista, it converts speech into wave and mp3 audio files too, all for free.

Hi,

Here's another new app that generates podcasts that sound like an audio documentary with sound effects and music in the background. The mp3 episodes generated by VocaTalk are in CD quality (44.1 kHz) stereo. It can use SAPI compatible voices and and digitally processes the voices to create a comfortable and fun listening experience. The more voices and music you have the richer the experience. Downside is, it's still in public alpha, but the author is looking for alpha testers and giving them full license if you volunteer.

http://www.vocamedia.com/
Edited by Omega Point, 30 July 2009 - 08:16 PM.

Oh I forgot to mention, VocaTalk can also put binaural beats to the generated podcast episodes. You can use alpha waves for relaxation, and theta waves for learning. With a good selection of music, it's really fun. I almost quadrupled the amount of information that get daily using this tool. The trick is, you don't even need time to read a book. It's just with you whereever you go.
Edited by Omega Point, 31 July 2009 - 04:55 PM.

There is a nice demonstration here -

http://sites.google..../highqualitytts

with very impressive quality.

I have also found a Text-toSpeech software,Power Text to Speech reader,it could run in winxp,vista or win7,contain several free voice.You could find it at http://www.1speechsoft.com

I use TextAloud for my Text-to-Speech Instance (you can save to .wav)
I use Accapella as engine for the voice Heather, her voice is pretty soothing when you turn down the speed in TextAloud.
Its nice for listening to blogs and e-books while I'm working.
Edited by Fleet, 22 March 2010 - 12:08 PM.

Has anyone used the software at www.myhousetalks.com?? It looks kind of interesting, and it's pretty cheap, but who knows how well it works? Thanks.

I found cheaper text to speech rit here http://www.speechtotextservice.com/

It works and free:
http://totitray.blog...ts-text-to.html

The synthesis still isn't great, but you may find some that are acceptable to your ears.

I've been using the UK voice on Natural Reader 7. To me the UK accent sounds most "normal", that's why I am using that one even though I'm American... you can test some out on their site: http://www.naturalre...s.com/index.htm

It's easy to copy paste text into it, and export to MP3, which I then export to my MP3 player...I had some manuscripts I listen to while I am working out (and also some articles from websites). Not great quality, but acceptable.

Also with that program, you can add music in the background (it merges it when it encodes the MP3, so the final MP3 can have music and speech), which I do at a low volume. Imo, it helps make it seem less annoying/artificial. I think it costed about $35.

 
Not very natural with what I made it read.

Hi, my name is Bruno. Chances are you're the nightguard. In that case, I'll have to go over the terms of service, because, well, you know, policies and stuff.