Synthesizer V Ver. 2: Koharu Rikka, Natsuki Karin & Hanakuma Chifuyu Reviewed

Popular AI singing voice databases receive a major overhaul

The Synthesizer V voice databases Koharu Rikka, Natsuki Karin, and Hanakuma Chifuyu have been updated to Ver. 2.
All three products are known for voices based on popular voice actors. They have now been optimized for Synthesizer V Studio 2, with additional singing recordings by the voice actors.
The main changes are as follows.

• New recordings and redesigned vocal styles make them feel like entirely new products
• Adjust a style's pitch, timbre, and pronunciation separately
• The gender parameter produces a much wider range of natural sounding voices
• Capable databases often used by professional composers for competition demo vocals

Product review video

▶ Watch videoWatch on YouTube ↗

Sample comparison by genre

We created samples for four different songs, adjusting each of the three characters' styles.
Each sample reflects the author's own taste.
Switch among A, B, and C and listen to these as examples.

J-pop

Rock

Chill pop

Idol pop

Starting with Ver. 2, you can also easily create doubling and harmonies like those added to the samples.

System requirements

These databases run in Synthesizer V Studio 2.
Synthesizer V Studio 2 is an editor for high quality AI singing synthesis, featuring an intuitive interface and fast processing.
It comes in Pro (paid) and Basic (free, with limited features), so you can start with Basic if you want to try it first.
See the following article for detailed instructions.

Updates shared by all Ver. 2 databases

Let's look at the changes shared by the three Ver. 2 databases featured here.

All three have new singing recordings by their voice actors. The pitch movement and breathing sound remarkably human.
This feels like a serious rebuild, not merely an update.
One point to note about Koharu Rikka: Ver. 2 is available only in an AI edition; there is no Ver. 2 Standard edition.

Expanded vocal styles

Vocal styles work like singing presets. Knobs in the Voice panel let you change the tone and nuance.
In Ver. 2, the number of styles has increased from five to seven. The additions vary by character, letting each cover genres that were harder to achieve before.

Ver. 2 also lets you adjust a style's effect on pitch, timbre, and pronunciation separately.
Pitch controls gestures such as scoops and vibrato; timbre controls vocal texture; and pronunciation controls clarity.
Each style has these parameters, giving you finer control over the sound you have in mind.

Improved gender parameter

The gender controls have also been updated for all three databases.
Previously, adjusting gender could make a voice sound artificial. This time it can sound more mature or more youthful while remaining like a believable real voice.
For Hanakuma Chifuyu in particular, gender adjustments can produce a voice that sounds more like a full fledged professional singer.

Faster processing

Synthesizer V Studio 2 has optimized multicore support, resulting in much faster processing.
Parameter changes take effect quickly, and the software runs without a high performance graphics card, making it comfortable to use on a laptop.

Koharu Rikka

The first TOKYO6 ENTERTAINMENT product was designed as a versatile voice for a wide range of genres.
Her voice actor is Yoshino Aoyama (Hitori Gotoh in Bocchi the Rock! and a solo singer); character design is by nari Teshima (character designer for Girls Band Cry).

New Mellow and Adult styles

Ver. 2 adds Mellow and Adult to the vocal styles, allowing for a more mature voice.

As you heard, adjusting Mellow and Adult creates a different expression from the default voice.
Combining Adult with the gender parameter lets you refine the sound further to your taste.

Using the Emotional style

My personal favorite for Koharu Rikka is the existing Emotional style.

Compared with the untreated voice, it changes both the projection and the nuances of breathing, adding a sense of forward motion.
This update expands the range of genres Koharu Rikka can cover, from relaxed chill music to emotional rock, with a single database.

Natsuki Karin

The second TOKYO6 ENTERTAINMENT product, Natsuki Karin, has an energetic, bright, and powerful voice.
Her voice actor is Miyu Takagi (Shinobu Inuyose in D4DJ), and character design is by nari Teshima.
The character was also chosen as the lead vocalist for an image song for Goddess of Victory: NIKKE.

Choosing among three Pops styles

In this update, Natsuki Karin's parameter setup has been substantially redesigned.
The earlier version was known for a cute voice suited to bright anime and idol songs. The new recordings focus on a cool tone, lower range, and rock expression.

Ver. 1's Happy and Falsetto styles have been removed and replaced by three Pops styles and Rock.
First, listen to the default voice without any style applied.

Before applying a style

Next are examples using each of the three Pops styles.

Pops is a less cute voice suited to J-pop; Bright Pops evokes a classic idol pop delivery; and Power Pops adds strength. The selection makes it easy to dial in the nuance you want, even within the same pop song.

How to create a powerful voice

The new Rock style produces an even stronger voice.

Combining Power Pops, Cool, and Rock can produce a highly emotional performance.
Cool gives a cool tone without Power, while Rock adds rock oriented dynamics.
If you used Ver. 1, the new result may be a voice you would not have imagined was possible.

What to do if sibilance is too strong

Raising Cool and Rock together can make sibilants (such as the Japanese s-series sounds) more prominent.
In that case, show the phoneme timingand reduce the parameters for sibilant phonemes such as “s” and “sh.”

Hanakuma Chifuyu

The third TOKYO6 ENTERTAINMENT product has gained users among Synthesizer V AI databases in recent years.
Her voice actor is Kaya Okuno (Verxina in Umamusume: Pretty Derby), and character design is by nari Teshima.
Unlike the other two, her voice is distinctly soft.

Subtle, expressive singing

Ver. 2 adds Pops, Mellow, Power, and Emotional—four new styles—expanding the range here as well.
Hanakuma Chifuyu's defining trait is her soft, breathy voice.
This database shines especially on gentle songs.

Adding Mellow to the default produces a slightly more focused, cool impression. Adding Adult as well can create a mature voice with a languid edge.

Pop versatility

Although Chifuyu is associated with gentle songs, Ver. 2's Pops, Power, and Emotional styles also make her suitable for pop and more forceful singing.

Even Pops alone changes her voice to a more assured, full bodied delivery.
If it still needs more punch, adding Power or Emotional produces a stronger vocal.
The Kawaii style also lets you adjust the nuance to suit cute or mature pop songs, making this a highly versatile database.
From mellow tracks to pop, this Ver. 2 database feels reborn.

What to do if sibilance is too strong

Raising Emotional and Power too high can make sibilants (such as the Japanese s-series sounds) more prominent.
In that case, show the phoneme timingand reduce the parameters for sibilant phonemes such as “s” and “sh.”

Feature comparison

Editor feature comparison

*1 More detailed AI Retakes are available.
*2 Pitch expression, timbre, and pronunciation can also be set individually.
*3 You can adjust duration and strength more intuitively below the piano roll.
*4 Supported only in ARA bridge mode.

Voice database feature comparison

*1 More detailed AI Retakes are available (only with Synthesizer V 2 dedicated voice databases).
*2 Detailed AI Retakes are available, but timing adjustment is not supported.
*3 Pitch expression, timbre, and pronunciation can be set individually (only with Synthesizer V 2 dedicated voice databases).
*4 Ver. 1 Synthesizer V dedicated voice databases cannot be used with Synthesizer V Studio 2 Pro.
*5 Richer vocal expression is available.
*6 The Ensemble series also supports switching chorus parts and randomly switching vocal variations.

Pricing

All three products are sold at the same price and can be purchased from the AHS official website.

• Boxed edition: ¥10,780 (tax included)
• Download edition: ¥9,680 (tax included)
• Upgrade from Ver. 1: ¥4,950 (tax included)

The required Synthesizer V Studio 2 editor costs ¥16,500 (tax included) for the boxed Pro edition and ¥13,200 (tax included) for the downloadable edition.
Basic (limited feature edition) is available free of charge.

For users who want the full experience, a starter pack with the paid editor and a coupon for one voice database costs¥22,000 (tax included), less than buying them separately.

* Prices are as of February 2026. Check the seller's website for current prices.

I've personally used Koharu Rikka since Ver. 1, and Ver. 2 felt like a different product from the moment I tried it.
The new recordings and redesigned styles in Natsuki Karin and Hanakuma Chifuyu also greatly expand the vocal range of expression.
One database can now cover many more songs. It feels less like using character software and more like hiring a capable vocalist.
If these sound interesting, give them a listen.

Frequently asked questions

Q1. What improvements do Synthesizer V Ver. 2 voice databases offer?

They add new recordings by the voice actors, increase vocal styles from five to seven, allow separate adjustment of pitch, timbre, and pronunciation, and make gender adjustments sound more natural. This expands the range of genres they can cover.

Q2. How much does it cost to start using Synthesizer V Studio 2?

There is a free Basic edition and a paid Pro edition. You can try Basic first without an initial purchase.

Q3. Can Ver. 2 databases create doubling and harmonies?

Yes. Ver. 2 supports doubling and harmonies, making them easy to create. The databases offer quality suitable even for professional demo vocals.

Q4. Can Synthesizer V Studio 2 run on a laptop?

Yes. Optimized multicore support improves processing, and you can work comfortably on a laptop without a high performance graphics card.