Blind and Visually Impaired Community
↑
3
↓
What kind of voice-synth programs do you folks use? (self.Blind)
submitted by Ol_Pappers
I'm not blind by any stretch of the imagination, but I have a fascination with voice synthesis / text-to-speech programs. My personal favorite is Dr. Sbaitso (an old DOS program, mostly for nostalgia purposes), but I also use Speakonia and I have emulators for the Atari ST and the Commodore 64 with appropriate speech ROMS. I also download non-English voice synth programs to periodically give my robot voices an accented flair. So far, I've only done this with Russian, but I can understand most of what he is saying.
What programs do you use? Do you prefer a Tee Tee Ess's with something that sounds more natural or do you prefer something that sounds a bit more robotic? Additionally, do you ever feel nostalgic for older voice-synth programs?
What programs do you use? Do you prefer a Tee Tee Ess's with something that sounds more natural or do you prefer something that sounds a bit more robotic? Additionally, do you ever feel nostalgic for older voice-synth programs?
fastfinge 3 points 6y ago
I used to use eloquence, but switched to $1. The problem with eloquence is that it's terrible for proof-reading. It's pronunciation rules are far too forgiving, meaning that if you make a spelling mistake, eloquence will say the word correctly anyway. I can always tell blind people who grew up using eloquence, just based on the spelling mistakes they make when they write.
Marconius 1 points 6y ago
Back when I had vision, I enjoyed messing around in text edit or other wordprocessors on the Mac playing with all of the novelty voices. Now that I absolutely rely on screen readers to interact with technology, I use voice over on my iPhone, iPad, and MacBook Pro. For this reason, I absolutely require that the voice is as realistic as possible and tend to use either Samantha, Ava, and occasionally Alex when doing testing as he has an extremely fast response time. The voice needs to be pleasant, and the internation needs to flow realistically and not be grating, and while novelty voices are fun, The more robotic voice the less useful and pleasing it is when using it over time.
-shacklebolt- 1 points 6y ago
I prefer Eloquence.
I've tried countless "natural" and "realistic" speech engines that I've found disappointing because they fall apart at higher speeds (or just don't go as fast as I want at all.)
Eloquence is most intelligible at high speeds (and supports really fast ones), consistent, and easy to listen to over long periods of time. I don't mind that it sounds somewhat "robotic" (less than espeak, more than many newer engines) and have been primarily using it for nearing a decade now (since I lost my sight.) The alternative I typically use is espeak if possible.
"Human sounding" engines can stick to GPS apps and bus stop announcements for me. For reading hundreds of thousands of words a week, I want power.
I've tried countless "natural" and "realistic" speech engines that I've found disappointing because they fall apart at higher speeds (or just don't go as fast as I want at all.)
Eloquence is most intelligible at high speeds (and supports really fast ones), consistent, and easy to listen to over long periods of time. I don't mind that it sounds somewhat "robotic" (less than espeak, more than many newer engines) and have been primarily using it for nearing a decade now (since I lost my sight.) The alternative I typically use is espeak if possible.
"Human sounding" engines can stick to GPS apps and bus stop announcements for me. For reading hundreds of thousands of words a week, I want power.
modulus 1 points 6y ago
Eloquence, which I think was developed by IBM and sold (or leased, or some weird business arrangement) to nuance. However Eloquence only does English, French, German, Spanish, Italian and Portuguese. Oh, and Finnish. For other languages, ESpeak is a good bet, though the quality of the voice is not as good.
I am not so bothered about the voice sounding natural as I want it to sound uniform, and to be understandable when it speaks very quickly. I also like voices which speak very precisely and make fine distinctions. So for example using Spanish voices (where most people don't pronounce v and b diffferently) I prefer the voice to do so.
I am not so bothered about the voice sounding natural as I want it to sound uniform, and to be understandable when it speaks very quickly. I also like voices which speak very precisely and make fine distinctions. So for example using Spanish voices (where most people don't pronounce v and b diffferently) I prefer the voice to do so.
Vaelian 1 points 6y ago
I use Vocalizer, which is a fairly natural speech synthesizer that comes bundled with MacOS and iOS.
The first text-to-speech program that I tried was Text Assist, a Windows 3.x program that came bundled with my Sound Blaster AWE32. It was a Notepad-like program that could read text out loud and even sing.
The first text-to-speech program that I tried was Text Assist, a Windows 3.x program that came bundled with my Sound Blaster AWE32. It was a Notepad-like program that could read text out loud and even sing.
This nonprofit website is run by volunteers.
Please contribute if you can. Thank you!
Our mission is to provide everyone with access to large-
scale community websites for the good of humanity.
Without ads, without tracking, without greed.
scale community websites for the good of humanity.
Without ads, without tracking, without greed.
©2023 HumbleCat Inc • HumbleCat is a 501(c)3 nonprofit based in Michigan, USA.