レトロさとりさんのレビュー一覧

This Text-to-Speech (TTS) software is the first one to feature the official English voices for the iconic Kotonoha twin sisters.

The voice quality does respect the high standard you might expect from A.I.VOICE, so that's a plus. Their voices sound natural for the majority of the time.

Unfortunately, there isn't much customization, as you mostly get the basic features you find in many TTS software. No emotion selections or anything like that.

Kotonoha Akane talks in a more down-to-earth style, while Kotonoha Aoi talks in a formal tone. Although, their voices don't resemble the original Japanese voices that much, so I suggest hearing those English voices through videos before considering the purchase.

Overall, the software has all the fundamentals but it lacks any special features that make a significant difference compared to the rest.

I wish the voices were more faithful to the original ones, but considering how this is the first English TTS software for the twins, I think it deserves at least a three-star rank.

Otomachi Una finally made her Voidol RVC debut. Currently, I find this RVC model to be one of the best in the Voidol RVC library.

With this model, Una is capable of singing and speaking in other languages other than Japanese, while keeping her signature voice.

You can adjust the pitch and the noise gate, as well as choosing between Low Latency, Balanced and Quality mode.

In Quality mode, the voice comes out with very little noise, but it takes a few seconds before you can hear the voice. Low Latency is the opposite: comes out much faster, but you will hear more noise than usual.

If you are a fan of Otomachi Una, you should absolutely consider a purchase.

Tohoku Zunko sounds pretty close to the original one thanks to the RVC technology used by Voidol RVC.

You can set a Noise Gate, the Pitch and the Conversion Mode.

Conversion Mode in particular, is useful depending on the situation: if you are doing a live stream for example, going in "Low Latency" mode is perfect.

Remember that Tohoku Zunko Voidol RVC will not work on Voidol 1, 2 and 3.

I bought the Talk Starter Pack just for Maki English alone, but I also tried the japanese one included.

Maki English reads pretty much anything you write in English without being forced to adjust settings. She does it with a clear voice too!

Maki English comes with 5 emotions: Fine, Angry, Calm, Sad and Mild. Her voice changes so much when you set different emotions. You can even mix them up together.

Maki English (as well as Maki Japanese) is a good choice for beginners and even experts.

This voice model works pretty well compared to some of the voice models I used like the Kotonoha sisters one.

Keep in mind that this is Voidol AI Voice Model, not Voidol RVC Voice Model: as such, it's not compatible with Voidol RVC. If you want to get some decent results, with all the AI Voice Models, you have to make some hardware preparations as well as knowing the ins and outs of audio configurations.

If you are a fan of the Kotonoha sisters, you should probably consider this voice model. It comes with both sisters' voice models, but none of them have any singing versions (still, I think they do have some singing compatibility with their normal voices but don't expect too much).

However, both voice models included require a lot of preparations. Advanced hardware and knowledge preparations. If you have all that, then this product is for you.

フィーちゃん's voice model is not a bad choice for the regular Voidol category. You can hear her original voice somewhat, but you need hardware preparations and knowledge if you want to get the best out of this voice model.

It also depends on which Voidol you use, so keep that in mind as well.

Last but not least, the model does have singing versions which is certainly useful for those users who want to sing with her voice (just remember that you still need to do lots of preparations just like with the normal version).

This voice model lets you talk in Tohoku Kiritan's original voice with almost-perfect accuracy.

It's no surprise that this voice model is such a good option, as it uses the RVC voice engine.

RVC is currently, for me personally, one of the best Voice Changer engines out there. Every voice comes out as close as the original character as possible. It also has a nice compatibility with different languages.

Don't expect too many options when it comes to expressing emotions, but humming should be fine.

Voidol 3 got an upgrade to Voidol 3R which now includes the Voidol RVC software. As far as I know, there is no other way to get Voidol RVC if not buying Voidol 3R which is pretty expensive right now.

If you can get past the price, then you are on a good track! Keep in mind that you need a good NVIDIA graphics card (maybe other graphics card could work, but please stick to what they said in the voice model shop description) otherwise the voice will not recognize the words properly and the software will struggle with the Quality option.

One more thing: if you hear noises in the background, you can try using filters on the final voice (not the source voice) and see if you can fix it up a bit. The Noise Gate you see in Voidol RVC is for the source voice, not the final one.

I used a Noise Gate filter on this voice and even with the minimal settings, the voice came out pretty good with minimal noise.

If you have Voidol 3R and want more voice model options, give it a go to Tohoku Kiritan Voidol RVC model.

This voice model is pretty stable as far as I tried. Even without the best settings, you can get some pretty decent results with it.

I tried this voice model with Voidol 3, which gives you the possibility to use Synth Mode as an addition to AI Mode (and viceversa). It really helps in making any voice more accurate to this voice model, so be sure to consider any newer Voidol softwares too.

With the right tools you can get some impressive results with this voice model but it's a nice choice to start with, in case you aren't sure to have the best equipment possible or expertise.

This product is perfect for those who want to offer their own personalized voice to use in a variety of digital activities such as Streaming, Video Production and etc.

Thanks to the feature of converting your voice in Syth mode first before going to AI mode, allows you to use a lot more voice models than before.

I tried almost all the 13 voice models included, by going with the Presets included, and they work pretty well even with non-optimal drivers and settings.

I used a Blue! Voice - Yeti Classic microphone for the test though, so I suggest to first be sure to have at least a decent microphone before considering a purchase.

Keep in mind that, in case you want to insert other voice models such as the Kotonoha Sisters model, you need to make a lot of preparations first. That means having the best drivers you can get (Voidol 3 is ASIO compatible), as well as having knowledge on how to make the best settings for Sample Rate, Audio Buffer Size and etc.

I'm not sure if all the additional voice models (to be purchased separately) need such preparations, but it's advisable to be well prepared anyway as, to this day, there is no way to try the additional voice models before buying them.

To sum everything up:

If you just need a voice software which gives you plenty of personalised voice options right off the bat, this product is for you.

If you are more interested in making videos based on the character voice model you buy separately and you want to be as accurate as possible with said character voice model (not such case for the 13 characters already included), then make lots of preparations with both hardware and software (as well as being an expert on making voices sound good) before considering a purchase.