blastbay studios tts voices

Hi everyone,

Thanks for sharing about the Blastbay voices — I agree that they sound very impressive, and it’s great to hear that Philip Benefall is planning to bring them to iOS, iPadOS, and macOS in the future.

I also wanted to highlight that another TTS product called Speechelo , which many may not realize, is also developed by the same person or team behind Blastbay , as mentioned in this thread. Speechelo uses the same neural-style voices, and being a cloud-based platform , it works across all systems — Mac, Windows, iOS, Android, and any browser. I personally subscribed to the Pro Plan , and the voices are quite expressive and natural, with multiple languages and tone options like friendly, serious, or joyful.

However, I do want to share an important caution for fellow screen reader users: Speechelo’s login page currently includes a visual CAPTCHA that is not accessible with screen readers . This means that as a VoiceOver user, I am completely blocked from logging in without sighted help. This is quite frustrating, especially since this is a paid tool that I use for work-related projects.

I’ve already reported this to their payment/support provider, Paddle, and I’m still waiting for a proper resolution. So, while the voice quality and platform independence are great, I strongly advise other blind users to be aware of this accessibility issue before purchasing.

I’ll post an update once the CAPTCHA problem is resolved. For now, please proceed with caution if you’re a screen reader user.

Best regards

Good morning. Just wanted to clear up some confusion. I am a close personal friend of the person who developed these blastbay voices. Speechelo was not developed by the same person. I’ve honestly never heard of that.

I loathe using SAPI 5 with NVDA, and as for Narrator, I use Jenny from the Neural voices.

Hmm, now if I could use these on my Xbox, then you'd have a winner. 🙂

these voices just sound flat and dull, just like microsoft Eva etc.

i'd pay if they were expressive like 11-labs but unless i am personally missing the attraction, they just seem uninspiring.

Great to see them out in the wild though.

Maybe the newer versions will have something but for now, I can get the same type of voices with sapi, maybe if you released them for free peple would download them but I don't think you'll get many peple paying for these.

When I tapped the play button on my iPhone, the voices weren't working. I rarely—if ever—have access to a computer, so I had to ask one of my cousins to pull up the website so I could hear what the voices sound like. I don’t know if they’re supposed to play on iPhone or if I’m doing something wrong, or if it’s that these voices are meant for Windows users. Now, I’m not trying to compare products because once I start doing that, I’ll always be chasing the next “better thing.” And while these voices aren’t as lifelike as some others, I’m hoping that if they do come to iPhone, and they’re offered as a one-time purchase, then yeah—I’d be totally on board. I really can’t afford a monthly subscription, but I love creativity, I love innovation, and I absolutely support people building tools that make a difference. I can’t do month-to-month payments. If there’s ever a lifetime option, or the ability to buy voices when I need them, then that would be ideal.

Also, knowing how locked-down Apple’s ecosystem is, someone can correct me if I’m wrong, but even if these voices came to iPhone, would we be able to use them for VoiceOver or Speak Screen? Personally, I’d love to use them to read books. That’s where the issue comes in for me. A lot of book platforms don’t have auto-page turning, their apps are hard to navigate, and they don’t include any built-in TTS or neural speech options. So even though Kindle isn’t perfect, it’s still the best experience I’ve had so far.

That said, if these developer voices ever do come to iOS, I’d definitely support them. I’d probably start by buying one voice first, then get the other one later when I can. And if there’s a bundle—maybe both voices for $25 or $30 total—I’d jump on that. Even $20 per voice would be fine. I’m always happy to support good developers, especially when I can hear the effort that’s gone into it. Even if I’m not blown away right now, if I believe in the product, I’ll still support it so it can grow into something better. There seems to be some confusion about the Speechelo voices. I’ve heard them too and I’m not sure if these are the same, but they’re definitely good. If voices like that ever came to iOS and could be used for reading books or even integrated into VoiceOver, and if I could buy them outright with no subscription? I’d absolutely go for it.

Anyway, I do hope the developers keep going strong, and I wish them the best. But also, can someone please start a thread about Apple voices? Because I’ve had Samantha as my default for years . We seriously deserve an upgrade. I feel like Apple could offer a voice package with new neural TTS voices—maybe even with a one-time cost per voice. I’d 100% pay for that.

you can do this since IOS 17 I think but as far as I know only espeak has bin released

Unless I am misunderstanding you, and apologies if that is the case, but you, can, buy these for $20, each.
Again, unless I am missing a bit of information here. 😣

Sorry for the confusion—I see that the voices are $20 each, and I definitely plan to purchase them once they’re supported on iPhone. I don't need them for my computer. I need them for my mobile device. I was wondering though, is it possible to purchase the voices directly from my iPhone and get a serial number, and then later transfer or activate them once they become available for iOS? Or would this not work?

Also, thanks for the info about eSpeak. I didn’t realize that this open source voice was available on the iPhone. And I see there’s a podcast too—I’ll check it out.

, bookmark the following Audio Games Forum thread:

https://forum.audiogames.net/topic/56672/new-release-text-to-speech-voices-from-blastbay-studios/

subject line says it all. I'd like to see English, welsh, Scottish voices in the offering. These voices sound good but region is important to me.

from my understanding the voices were trained on audio data, so it would be really cool to have tonnes of UK voices: Scottish, Irish, Welsh, jamaican, African, cocny, come on, who'd not want a cocny voice, brummy/burmingham, black country, northern, scouser, liverputlian, I'm sure I'm missing some but yeah, if you could make it so we could train our own voices on data, perhaps from the internet archive, i'm not sure, that would be a standout feature, start with a blank slate and go from there.

At the moment there's no way I'd pay $20 for these voices, they don't do anything the every day sapi voices can't do.

These voices do sound rather robotic, although robots now sound like people... There's also (or was) an experimental neural voices addon for NVDA that accessed some Microsoft voice models, though not the ones used by Narrator. My impression of the thread discussing the addon is that Microsoft has some freely available.
I swap between Eloquence and George, but flip over to Narrator Ryan or Jenny when I want my writing to sound better than the voice in my head. They were jaw-dropping when they first came out. Now, to put my jaw on the floor, I hit control-alt-space in Word for it's read aloud feature. I swear, I used to work with that woman. She was cute. A monster, turned out, but cute. Now I make her proofread my manuscripts and there's nothing she can do about it bru-ha-ha.

Subject says it all. I really like the Jenny neural voice for narrator, and would love to have that for NVDA. I would mention aria, but I think aria is a little broken. For some reason, whenever I try to load her up on narrator, she has stuttering issues.
No, seriously…
But Jenny, and yeah Ryan too, those two are great. In fact they can replace David and Zira, any day now. 😇

I poked into it, and there are several initiatives to bring neural voices to various platforms. There's a Sonata addon for NVDA that uses free Piper open source voices, a couple of which are somewhat expressive. I'd use them, except that the archive page for getting the voices seems to be broken. Anyway, it's all coming round the mountain, seems like. The OP's product will doubtless receive more training and improve like the others. That being said, they'll have to pry Eloquence from my cold dead hands for the most part.

it's odd isn't it? You'd think we'd want more expressive voices, but those of us who use eliquence don't,, "ok, you say, switch to ESpeak, surely that's the same thing?" But it's not, ESpeak sounds robotic, Eloquence blends that sort of vaguely humanness factor with a machine like sound that's not to grating, or at least for me.

For eSpeak, it depends on which variant you use.

I personally use the Robert variant, with inflection all the way up at 100 and pitch set to 40.

For the longest time, I used to use Eloquence, but now I only use it if I don’t have any other choice.

On my phone, I use Fred.

Fred sounds fine on a phone speaker, but on a larger speaker, it sounds a bit weird.

It is possible to use the neural voices including Jenny for NVDA. All you need is the Neural Voices addon for NVDA. You can find it on Jake Gross' Website, datajake.braillescreen.net. Go to tts, synthesizers for nvda, and you will find it.

It doesn't work with the latest version of NVDA, or at least it didn't for me.

I will check that website out, and get back with you on that. As it stands right now, both my iPhone, and my HP laptop, are running Vocalizer Samantha. I would kill for a working Jenny, however, for my laptop. 😍

I'm using this tool: https://github.com/gexgd0419/NaturalVoiceSAPIAdapter

All you need is to download the last working versions of the voices, rename them to zip, unextract them into a folder and point the adapter to the path where you extracted the voices. No problems using the offline natural voices for me. All you need is in the link above, including the last working versions for the voices.

It’s just like my subject line size. :-) I think the reason that we are not yet seeing these kinds of voices on screen leaders is because they have to be able to read so much information and they have to be able to keep up with the amount of varying information on screen. There’s a lot to process in the computer. Even the Voice is inside of our phones when they are at the enhanced premium qualities are definitely suffering. This is why we had the bugs with the phones constantly crashing with the screen curtain on when it would read with lots of Text when the first premium boys came to be in iOS 16. And these are not even the premium high versions of vocalizer voices. They are stepped down from those. These are the older ones that John used to give us. These are the older ones that were called premium high. Then the variant below those is the enhanced ones that were just available for I believe iOS iPadOS. I don’t use a Mac so I don’t remember exactly what’s available in those. I know there’s compact and available. John has this as well. So basically we just don’t have the processing power to handle it. When we were given all the voices that we were given in iOS 16, that’s what my phone kind of went nuts. That’s why we’ve gotten so many basic voices that came preinstalled such as eloquence and the apple smaller mass and talk voices. Although there are four of those that require downloads as well, such as Fred Preinstalled, Vicky, Victoria, Agnes and Bruce are all downloadable. I don’t know why those didn’t come for himself since they’re so tiny. But not the ones that are preinstalled is Cathy, Fred, Junior, and Ralph. I don’t know why and I’m probably sure it’s because it requires more data, but there are no enhanced or premium that are preinstalled. They all need to download. Apple explained this to me once as the voices that need more data of the ones that have to be downloadable. I have just recently as in three weeks ago became a hearing aids user and now I have to use an enhanced premium voice. I get more clarity out of hearing those voices because I do have better addiction and pronunciation. So right now I have to use my favorite Voice, which is Karen, but I have to use her premium variant as I said due to clarity of addiction and pronunciation. I didn’t mean to make this a really long post and I’m sorry about that. But I believe this is why you did not hear the more natural voices in windows being used on any screen readers at this point. I know those things in Microsoft that have more natural voices and all that but I still don’t see how something that would be that responsive, but necessarily will be processed correctly yet. Also, as much as I love voices and screen readers, I am really not willing to pay for voices. I have enough things that I pay for as it is, is it subscriptions that I have Enough things that money that’s going up every month or even a one time amount is just too much for me right now. I hope this helps out somebody though and I hope that you know anybody that likes these new voices will enjoy them.

Hi, I wanted to at least get to hear these voices. I’m not going to buy any voices at the moment and I don’t have Windows 11 because my computer cannot process that. Will run Windows 10 using jaw 2025? I do not yet have NVDA installed. I’m planning on doing that at some point though. I wanted to at least hear what these voices sound like. I love the names, Richard and Libby. I wanted to just hear the samples and there’s a play button, but when I double tap it nothing happens. It’s the case for both voices. Thanks for anybody’s help with this. :-)

I'm not a fan of the subscription model. I understand a lifetime subscription can’t be an option for every app. If you pay a lifetime subscription for $20 for, let’s say, a generation app, when customers buy that lifetime subscription, how do the developers get paid afterwards? You pay for that one lifetime subscription, and then they have to keep developing new features - video generation, image generation, avatar creation, file converters, PDF readers. They can’t give you the product forever without another income stream. If we don’t have monthly or even yearly subscriptions, they would need a secondary income to be able to support themselves and to also keep developing the app. If there are premium features available as a one-time purchase, I'd definitely go for those. For example, if there's a package of five or more voices, I might not be able to afford them all at once, but if I can buy them individually as a one-time purchase, I can save up and get them later on.
When it comes to subscriptions, unless it’s for an app I need professionally, I tend to rotate what I subscribe to. One month it might be Kindle Unlimited, the next month Cobo (which is actually getting more accessible, by the way), another month Netflix or Hulu. I try to keep it to one or two active subscriptions at a time. I guess they could have the lifetime subscription for the features, but still keep the advertisements for revenue outside of that aspect. I don't know how many individuals would pay for this model, since when you pay, you generally want to get rid of the advertisements, especially the ones that don't play well with voiceover. If I can manage without, I do, because I can’t afford to pay for everything monthly.
I totally understand developers need to earn a living, but I also have to be smart about how I spend my money and what I’m willing to prioritize. Rotating subscriptions is how I make it work. If I need to finish a project, I try to get as much done on the payed plan. If it's not a necessity, groceries, bills, finance, mobility, then I can live without it. I have to say, the Samantha default voice on my iPhone is the one I use the most, and I have no problem with it for VoiceOver on my phone. However, it would be really nice to have more natural-sounding voices specifically for reading . That's the only reason I'd want extra premium voices—not for daily iPhone tasks, as I don't need an emotionally responsive voice for this. When I'm using my phone from a general standpoint, I like having the robotic neutral cadence of voiceover. For anything creative, especially when it comes to books, I've been experimenting with downloading voices to read with the Speak Screen feature, But whenever I crank up the speed, the voices tend to get very high-pitched, almost like a chipmunk , and I can't understand anything, even with the premium voices. It feels like on the iPhone, Samantha default is the only voice that consistently works well and is designed to handle those higher speeds. When I try other voices, even the premium ones, once they hit that four-times speed, they start crackling and sounding too fast; choppy, weird static, the works. I've also been trying out Microsoft Edge's read-aloud feature on my phone by uploading and opening files within the app. The voices there are pretty good, but they don't work with web pages, especially for reading books, probably due to copyright. If I have a PDF or EPUB file of what I want to read, it works well, but it still sounds very choppy . Another issue is that when these voices are reading, they often seem to insert extra line spaces where they shouldn't be, which makes it hard to use. The Microsoft Edge browser on the phone definitely has some decent voices, and while I wish I could download them for VoiceOver or directly onto my iPhone, I understand why that's not possible.

The samples will only work with SAPI 5 running on your device. You mentioned Windows 10? That should be fine to test out the samples for these voices.

HTH. 🙂

https://www.youtube.com/watch?v=z_QOFlcm6vg

, is where ya can listen to it in full.

NB. Said Blind Android Users podcast in general, is laid-back & chill. Thus, get your popcorn!