Takayuki Ryou (隆行遼) is humanoid robot of unknown function and origin. Their name means “moving to distant heights”. They fly through space, alone, traveling from one planet to another. Their intention is unknown, however they have been known to sing. Is this something they were programmed to do? Perhaps they came from or have been engrossed in a society where its inhabitants sing to each other? Ryou’s voice seems to calm those around it, as well as themself.

隆行 (Takayuki) means “moving to heights”. 遼 (Ryou) means “distant”. Together, their name means “moving to distant heights”.


NAME/名前: Takayuki Ryou (隆行遼)
GENDER/性別: Xジェンダー (masculine presenting/sounding)

PRONOUNS: They/them, he/him
AGE/年齢: ???
HEIGHT/身長: 193CM
WEIGHT/体重: 150KG

Illustrations by Oscar Sprangers

Takayuki Ryou’s voicebank is a cool-toned Japanese voicebank. The lower range is husky and soft, whereas the higher range is sharp and powerful. This voicebank also includes vocal effects like breaths, vocal fry, and glottal stops.


BREATHS: Ryou’s voicebank contains several different inhaling samples, as well as breath-endings. Inhales can be activated by labelling a note as “br”, “br1”, “br2”, and “br3”, while breath-endings are activated by labelling a note as “H”.

GLOTTAL STOPS: Glottal stops are also included in this voicebank. These can add a bit of edge to the voice. These samples can be activated by adding a “GS” marker to the note’s label. (eg. “GSa”, “GSe”, etc)

VOCAL FRY β: Ryou’s vocal fry can add even more edge to the voice. This can be activated by adding a “VF” marker to the note’s label. (eg. “VFa”, “VFe”, etc)


This voicebank is based off of over 1000 configured samples, which allows for a smooth, natural-sounding synthesized voice.

Thanks to DeepVocal’s advanced synthesis engine, as well as the tightly-configured vocal samples, Takayuki Ryou is able to sing in a wide range of speeds. Takayuki Ryou is a flexible, well-rounded voicebank.
DeepVocalの高度な合成エンジンにより、隆行遼は幅広い速度で歌うことができます。 隆行遼は、柔軟でバランスの取れた音声ライブラリーです。

Takayuki Ryou’s Japanese voicebank is recorded and configured at three separate pitches. One at B2, another at E3, and the last pitch recorded at A3. This gives the voicebank a wide vocal range.
隆行遼の日本語音声ライブラリーは、3つの別々のピッチで録音されます。 最初はB2、2番目はE3、最後はA3です。 これにより、音声ライブラリーの音声範囲が広くなります。

DeepVocal is a vocal synthesis program developed by boxstar and publicly released in beta in 2019. The DeepVocal suite of programs allows users to develop their own vocal database for use within the DeepVocal editor. The DeepVocal editor is a composition program used to compose vocal tracks.

One might be inclined to use a vocal synthesis program as opposed to a human vocalist for a number of reasons. DeepVocal produces impressive results and provides an easy-to-use interface. Learn more at www.deep-vocal.com.

Takayuki Ryou’s Japanese voicebank was developed for the DeepVocal synthesis engine for use with the DeepVocal Editor. You can download it here.


Takayuki Ryou’s Japanese voicebank is scheduled for release in 2020. More information to come soon.

*This sample is not a demo and is therefore not representative of the final product.