The biggest flaw in my studio (aside from the lack of a serious musician and sound engineer ) is a singer who allows me to avoid putting my poor voice in front of a microphone. So I tried Suno, which I liked for its female Arabic vocals, but Suno is distracting with its “personality,” which tries to impose unexpected styles and notes. So I tried Synthesizer V Studio 2 Pro. Its only flaw is the language pack, which only contains English and Asian languages (Portuguese coming soon… and I like that). I tried this plugin with a song by my favorite Englishman… Mr. Sting. My ignorance of the English language made it very difficult for me to separate the syllables, and I’d like the opinion of a native speaker. So I quickly recorded “The Shape of My Heart”… I apologize in advance to Mr. Sting, whom I love and whom I had the pleasure of hearing at the Verona Arena many years ago.
As usual, your work is very well played and recorded. Some critical feedback for you:
On what hardware and OS did you run the plugin? It sounds fine. I’m not keen on the voice but that is personal preference. The voice sounds like it is autocorrected which looses some authenticity. The song is mostly played and recorded well with the voice laid well into the mix. The instrumental break at 2:30 is a bit too tight. There could be some movement around the beat.
Thanks @riban for the feedback. I recorded as usual with the MacMini M4 and Fender Studio Pro (the upgrade from Studio One 7 was free, but I kept the latter because it has features that are important to me that Fender doesn’t have, such as the ability to switch from 4/4 to 5/4 on a song that requires it). I’m going to listen to the part you mentioned right away, but I’m not sure if your comment was about a vocal or musical part. What about the pronunciation? Is it accurate for a native English speaker? Thanks.
There’s no automatic correction. I played the vocal notes on the keyboard following a score and added the voice to the notes… I didn’t quantize because the voice has notes advanced and delayed by the score.
Thanks @riban, I was really interested in this, given my excellent knowledge of English . I was afraid of errors in the separation of syllables because to divide a word on (for example) three notes, you have to insert like this: “Natural + +”, while to extend a word on the next note you add (space and -). The software supports English and many Asian languages; Chinese is the default, and on my first try, after entering a few sentences in English, I realized they were unlistenable even for me. Then I realized I was having a Chinese singer sing in English… Heartbreaking! I’m eagerly awaiting the Portuguese upload, given my previous experience with a Bossanova group.
The Synthesizer V 2 Pro Beta update has added new languages: German, French, and finally PORTUGUESE!!! Fantastic!
Here’s an example of the first take on a wonderful piece. The vocals were added without (yet) paying much attention to meter, but I think you can hear the potential.
Thanks @tunagenes, I appreciate the like. I’ve just replaced the track because I realized I had uploaded an old save file where both vocal parts were identical—meaning the second (more poetic) part of the lyrics was missing… Old age strikes again!
I am (almost) happy to announce that the new beta version of Synthesizer V2 Pro includes the Italian language… although the pronunciation is more or less like mine when I try to speak any language other than my own.
I should add that all the work is my own… AI played no part in the project.
Synthesizer V, as it stands now, it’s not very useful for my own specific goals, not being integrated directly in any music engraving program.
As you know, I have mostly a notation-based compositional approach, therefore visual and requiring a comprehensive view of what is going on in all the musical parts.
SV2 can be used, and it’s actually employed by many, inside notation apps, but (currently) only as an added tool separated from the score view: not really up to my tasks, too toilsome and too much extra tinkering, if one like me mostly writes for several voices at once.
While I acknowledge than SV2 is already way more convenient than other similar plugins available (and, believe me, I almost know them all ), its vocal models tend to not suit my taste and needs, since most of them are song/pop oriented or classical but in a strongly lyrical sense.
I absolutely agree with you, @Aethermind. For me, too, this software serves as a stopgap for the complete lack of decent singers in the area where I live. I have a folder on my PC containing lyrics I’ve written over the course of about 10 years—lyrics that aren’t really suited to becoming songs, yet they possess a certain poetic quality (humbly speaking, I’ve even won a poetry prize). With this software, I can input my lyrics exactly as I envision them. I’ve just written a funky/rap track that might actually work (in ITALIAN!!!). I’m well aware that human voices can’t be compared to software—which, incidentally, still requires a great deal of work to get even a passable result—but for an amateur looking to make use of lyrics that had been all but forgotten on an SSD, I think it’s a huge help. In fact, for this new track, I’ve been searching for a young female singer to record the vocals… though, so far, without any luck. I’d love to use the Royal Philharmonic Orchestra to arrange a track, too, but unfortunately, I have to settle for simple VSTs. My enthusiasm for this software stems from the fact that I can hear what it sounds like to have a decent voice singing my lyrics, all without having to pay a professional—who, quite rightly, would bring their own personality to the performance, which might not be what I’m looking for. I’ve attached one of the tracks I created using this software.
That is certainly true, and will be so for quite a long time, maybe forever.
The whole point is having simulated singers loyal to the written music, and with a neutral kind of interpretation, as average as possible, that the composer needs to make his work really shine.
Sometimes, to be honest, an excess of personality of the singer is a serious problem for the composer and the music. Some performers tend to try and take possession of the score, so to speak, imparting their idea of the work and shading on it a light on their own.
Some other performers don’t, and therefore they are those who are professionally most sought-after.
Virtual singers based on vocal LLMs generally belong in the first category of narcissistic performers: that is why I almost never like what they would bring to my writing. With pop/song singers from SV or Ace Studio this tends to happen practically always.
I disagree with you on this, @Aethermind. If we’re talking about Suno, then sure—but with Synthesizer V2 Pro (provided you have plenty of time to dedicate to expressiveness), you’re the one inputting the notes and tweaking the phonemes—lengthening or shortening them, letter by letter and syllable by syllable. For every single letter and syllable, you can adjust the delivery: make it more “powerful” or more breathy, add dynamics, apply vibrato, or keep the note steady. The catch is that it requires a massive amount of work—A professional singer wouldn’t make you waste all this time… maybe.. Below is an example of one of my tracks: I fed the vocals—created in English using Synthesizer V2 Pro—into Suno and asked it to recreate them, keeping the same timbre but in Italian. Naturally, it added keyboards, bass, and drums, all of which I promptly discarded.
) is a singer who allows me to avoid putting my poor voice in front of a microphone. So I tried Suno, which I liked for its female Arabic vocals, but Suno is distracting with its “personality,” which tries to impose unexpected styles and notes. So I tried Synthesizer V Studio 2 Pro. Its only flaw is the language pack, which only contains English and Asian languages (Portuguese coming soon… and I like that). I tried this plugin with a song by my favorite Englishman… Mr. Sting. My ignorance of the English language made it very difficult for me to separate the syllables, and I’d like the opinion of a native speaker. So I quickly recorded “The Shape of My Heart”… I apologize in advance to Mr. Sting, whom I love and whom I had the pleasure of hearing at the Verona Arena many years ago.
.