Onseogosei no Seiyujimusha Celebrates 1 Year Anniversary with Singing Synthesizer!

While Onseogosei no Seiyujimusha, translated as “Voice Synth Voice Actors’ Office” is far from new, they have a rather new addition that’s bringing the series of voice banks to new levels! In addition to their many talking voice banks, they now have a singing voice bank unlike any other.

About Onseogosei no Seiyujimusha

As mentioned, Onseogosei no Seiyujimusha translates to “Voice Synth Voice Actors’ Office.” It’s a series of voice banks with a little bit of plot surrounding them. These banks are all members of an office and work to serve those who hire them.

Their introduction even sounds like an office secretary wrote it!

“‘Onseogosei no Seiyujimusha’ is a voice synth project that started in 2017 and is celebrating its first anniversary come this spring. Our voice synth software, which is capable of expressing emotions, will work hard to fit your needs!”

About the Software

Onseogosei no Seiyujimusha utilizes the VoiceText engine by HOYA CORPORATION. Unfortunately, at this time, it’s only available commercially. However, there’s always a chance that this will change in the future!

The software’s official page advertises four special features of voice text: Emotion, Making Original Voices, Changing How the Voices Talk, and the Singing Synthesizer. The following points are translated from the website. Please note that the translations have been changed slightly to make more sense in English.

Point 1: Emotion

With VoiceText, happy, sad, and angry emotions are available. Through expression settings, you can control the strength and weakness of the voice. Through the emotions settings, you can control the strength of emotions. This allows you to have the phrase sound however you’d like it to. This makes for fun conversations featuring voices that don’t sound like robots but like humans.

Point 2: Making Original Voices

You can create your own original voice synth by recording your own voice or the voice of a voice actor. You can even revive the voice of people who have passed away. In addition to creating the voice itself and giving it emotions, you can attach special sentence-enders to all the sentences (like a cat ending every sentence with “Nyan!”). You can find an in-depth tutorial (in Japanese) here.

Point 3: Changing How They Talk

You can combine a character’s voice with another character’s speaking style. You can combine characters with different accents to create new ways of speaking or combining two voices to make them cuter. Perfect for people who want a different voice from those provided.

Singing Synthesizer

You input the melody and lyrics, and create a song. Unlike other vocal synths, it doesn’t go by one sound at a time but instead uses a statistical model to create smooth and clear singing. It has the warmth of a human being singing.

They are using a new style of singing voice synthesis technology developed mainly by Professor Keiichi Tokuda of Nagoya Institute of Technology, and speech analysis and synthesis system WORLD developed by Professor Masanori Morise of Yamanashi University.

The Characters

At the moment, there are 11 characters capable of speaking — one of whom is debuting as an idol. Unless otherwise stated, the voice is capable of happy, sad, and angry emotions.

The Newcomer Idol

Maid: “Wake up, it’s morning♪”
Idol: “I finally get to make my stage debut!”

Name: Hikari
Gender: Female
Age: 21

The clear voice of a female maid that’s good for news reports. The first voice made to express emotions.

Normal Voice Sample

Singing Voice Samples

The Reliable Butler

“You called, master?”

Name: Takeru
Gender: Male
Age: 21

The voice of a male butler who exists to work alongside Hikari. He has a very kind personality, and a calm and eloquent voice that’s good for news reports.

Normal Voice Sample

The New DJ

“I want to liven up your boring everyday life with my voice.”

Name: Shougo
Gender: Male
Age: 25

This character has no emotion options.

The voice of a passionate and hard working DJ. Suitable for entertainment purposes.

Normal Voice Sample

The Energetic Girl

“I’m gonna work really hard to earn my keep!”

Name: Hina
Gender: Female
Age: 8

The voice of an innocent 8-year-old girl. She has a clear distinction between on-the-job and off-the-job, and is a very calculating person who works hard and doesn’t mix business with pleasure.

Her voice is best suited for usage in conversation apps and robots.

Normal Voice Sample

The Kind Boy

“I’m almost done with my homework, can you wait a moment?”

Name: Kouta
Gender: Male
Age: 8

The voice of a really energetic 8-year-old boy. He’s very serious about greeting people properly. His best usage is for conversation apps and robots.

Normal Voice Sample

The Bilingual


“How do you say ‘Come pick me up’ in Japanese?”

Name: Emma
Gender: Female
Age: 12

Capable of Japanese and American English.

A female voice who can speak English. She’s bright and cheery, but also saucy and cheeky. However, she can’t read situations at all. She currently lives in America and she’s half Japanese and half British. Her voice is good for bilingual purposes.

Normal Voice Sample

The Genius Cat

“There’s no such word as ‘impossible’ in MY dictionary, nyan!”

Name: Tuna (Tsuna)
Gender: Male
Age: Unknown

A cat-like voice that always go “Nyan” at the end of his sentences. He’s very bright and easily elated. However, despite looking really cute, he’s actually a very smart cat. His voice is suitable for knowledge apps.

Normal Voice Sample

The Star of Our Office

“Did you all watch my show last week?”

Name: Show
Gender: Male
Age: 40

This character has no emotion options.

A male voice bank with a very unique and light way of speaking. He appears to be a hard-working kind of guy but, in reality, he’s really good at poking fun at people and making retorts. Everyone’s senpai. He’s been working in the office longer than everyone else. His voice bank is aimed at entertainment like being a TV show announcement host.

Normal Voice Sample

The High School Girl from Hakata

“When I’m with you, anything and everything is fun!”

Name: Kanna
Gender: Female
Age: 17

A very calm and serene voice. Kanna is a very serious and hard-working person. She’s very bright and friendly, but has a stubborn and pig-headed side as well. The character on her phone charm is called “Ikementai-kun.” Its name is a combination of “ikemen” (hunk) and “mentaiko” (pollock roe). Her voice is suited for any kind of voice guidance and Hakata-accented voice usage.

Normal Voice Sample

The High School Girl from Osaka

“It’s been a while, hasn’t it? How’ve you been doing?”

Name: Nao
Gender: Female
Age: 17

This character has no emotion options.

A voice bank with a brisk and energetic Osaka accent. Nao is a fun loving personality with a lot of male and female friends alike. But she’s also scatterbrained at times. She likes to root for her favorite baseball team with a megaphone that has the character “Takoyan” on it. Her voice is suited for any kind of voice guidance and Osaka-accented voice usage.

Normal Voice Sample

The Cute Robot

“I come programmed with state-of-the-art AI!”

Name: Robota
Gender: ?
Age: ?

A very clever, cute, and calming voice that’s good for calming people down. He’s equipped with artificial intelligence but, as one would expect, his true ability is…? His voice is appropriate for any kind of robot’s voice.

Normal Voice Sample

More Information

For more information, be sure to stay tuned to VNN! However, you can also check out these websites. The Onseogosei no Seiyujimusha website has voice samples of all character emotions, so be sure to have a listen!

Onseogosei no Seiyujimusha Official Website
VoiceText Official Twitter
Voice Text Official Facebook

About Hoshi

She's a huge VOCALOID fan, a figure collector, and an all around nerd. Hoshi's been in the VOCALOID fandom since 2008 and has been a hardcore VOCALOID fan since 2013.