This page serves as a way of documenting all the files included in the Artists/RetroBunn folder in the DECtalk Archive. This page includes comments on how the files were made, and what software was used to create them, as well as credits for any third-party material used (such as instrumentals).
The first file uploaded to this folder, this work-in-progress cover of the song was created using Natalie, a voice for Synthesizer V Studio 2. This takes advantage of the ability for voices to sing in five supported languages, regardless of what the native language of the voice is.
This rendition of the song is based on the arrangement of the song by Japanese singer Ken Hirai, using verses from both the Japanese and English versions of the song.
This work-in-progress cover of Tsubasa (translated as Wing in English), is created using Synthesizer V Studio 2 by Dreamtonics Co. Ltd, a singing synthesizer that is similar in functionality and use to Vocaloid, but it generates vocals using a different approach that is still human-sounding.
The song comes from the Sound Pool series, a series of music packs for creators and producers, the packs being created by AH Software Co. Ltd, an importer and seller of various music software in Japan. The song would be reused several times for showcasing their Vocaloid and singing synthesizer products. The vocals used in this cover are provided by a Japanese female voice known as Haruno Sora, also developed and distributed by AH Software.
This work in progress cover of Country Road was once again created using Synthesizer V Studio 2. The vocals were provided by Ayame, a Japanese female voice. The karaoke instrumental is by ACOUSTIOKE on YouTube.
Translated as Lana the Little Witch in English, this song was created using Synthesizer V Studio 2. The vocals were provided by Natsume Itsuki, a female Japanese voice created by Hifumi, Inc. based on the singer of the same name.
This song is also part of the Sound Pool series much like Tsubasa was, and much like that song, it was used a couple of times as a demo song for a few Vocaloid and later Synthesizer V voices from AH Software.
Not to be confused with the Hatsune Miku song of the same name, this work-in-progress cover of the song is once again made with Synthesizer V Studio 2. The vocals were provided by Gumi, more specifically her Synthesizer V version.
Originally a Japanese song, the English version of the song (which this cover is based on) was created back in 2013, and was used as a demo song for Gumi's original English Vocaloid3 vocal. The lyrics in this version were changed so that they sounded a little better in terms of how the lyrics flow together.
This song, originally performed by Laufey, was created using Synthesizer V Studio 2, using Natalie once again.
This interpretation of a classic Christmas song is once again created using Synthesizer V Studio 2. As the name of the file implies, this was created with twelve voices in a single project file.
The arrangement of the track is based on the Pentatonix arrangement. I only decided to cover the first part of the song, as the rest of the song would've been too complex to work on. The table below shows which voices were used in the cover.
| Vocal Part | Voice Name | Voice Gender | Native Language |
|---|---|---|---|
| Soprano (main vocal) | SOLARIA | Female | English |
| Soprano (first chorus) | Sheena | Female | Japanese |
| Soprano (second chorus) | Ling Wan | Female | Cantonese Chinese |
| Alto (main vocal) | Natalie | Female | English |
| Alto (first chorus) | Miki | Female | Japanese |
| Alto (second chorus) | Eleanor Forte | Female | English |
| Tenor (main vocal) | Kevin | Male | English |
| Tenor (first chorus) | Liam | Male | English |
| Tenor (second chorus) | Riku | Male | Japanese |
| Bass (main and chorus) | ASTERIAN | Male | English |
This rendition of a classic Christmas song was created with Synthesizer V Studio 2, and it uses a combination of English and Japanese voices. The instrumental was provided by Christmas Songs and Carols - Love to Sing on YouTube.
The table below shows which voices were used for each verse.
| Verse | Voice Name | Voice Gender | Native Language |
|---|---|---|---|
| Partridge in a pear tree | Natalie | Female | English |
| Turtle doves | Riku | Male | Japanese |
| French hens | Felicia | Female | English |
| Calling birds | Hayden | Male | English |
| Golden rings | Asumi Ririse | Female | Japanese |
| Geese | Ryo | Male | Japanese |
| Swans | Natsume Itsuki | Female | Japanese |
| Maids | Kevin | Male | English |
| Ladies | Haruno Sora | Female | Japanese |
| Lords | SAROS | Genderfluid non-binary | English |
| Pipers | SOLARIA | Female | English |
| Drummers | Ritchy | Male | English |
This arrangement of a famous traditional Christmas carol is performed by Asumi Ririse, a Japanese female voice for Synthesizer V Studio 2. The instrumental was provided by Stingray Karaoke on YouTube.
This rendition of a popular Christmas carol is created using Synthesizer V Studio 2. As the name implies, this was created using four individual voices; one for each verse. The instrumental was provided by Sing2Piano | Piano Karaoke Instrumentals on YouTube.
The table below shows which voices were used in each verse.
| Verse | Voice Name | Voice Gender | Native Language |
| First Verse | Ayame | Female | Japanese |
| Second Verse | Hiyama Kiyoteru | Male | Japanese |
| Third Verse | GUMI | Female | Japanese |
| Final Verse | miki | Female | Japanese |
Another cover of a traditional Christmas carol, this rendition was created using Synthesizer V Studio 2, but with a twist.
This combines four languages into one project, using Japanese, English, Mandarin Chinese, and Spanish versions of the song, with each language having their own verse. The final verse of the song wraps it all up by combining all four languages into a single verse. The instrumental was provided by Sing King on YouTube.
As the name of the file implies, this performance was done with fourteen virrtual singers in one project, who are all female voices.
The table below shows which singers were used.
| Voice Name | Native Language |
| Asumi Shuo | Japanese |
| Amara | English |
| AVER | Japanese |
| Eri | Japanese |
| Hibiki Koto | Japanese |
| Mai | Japanese |
| Miyamai Moca | Japanese |
| Mo Xu | Mandarin Chinese |
| Nakuru | Japanese |
| Nekomura Iroha | Japanese |
| Otomachi Una | Japanese |
| ROSE | Japanese |
| Tsurumaki Maki | Japanese/English |
| Yi Xi | Mandarin Chinese |
The first submission for 2026, this cover of Elvis Presley's Can't Help Falling in Love was performed with two different singing synthesizers.
The male vocal was created using Synthesizer V Studio 2. The voice used is known as REMMANT, a voice described as an old-school vocal designed primarily for jazz and oldies usage.
The female voice was created using VOISona, a singing synthesizer developed by Techno-Speech, Inc. The voice used is known as IA, a female Japanese voice originally released for VOCALOID3 back in 2012. This cover uses her English vocal made for VOISona, allowing her to sing in English.
The instrumental was provided by Mega Backing Tracks on YouTube.
This is a work-in-progress cover of I Only Wanna Be With You by Dusty Springfield. This was created using Synthesizer V Studio 2.
This arrangement uses two vocals. The main lead vocals were provided by Natsume Itsuki, a Japanese female voice who is based on the Japanese singer of the same name. The harmony vocals were provided by Nakuru, a Japanese female vocal, also based on a Japanese singer who shares her name.
The instrumental is taken from the Samantha Fox cover of the song, and was available on her "I Wanna Have Some Fun" album.
This is a sample of all of the MacinTalk voices included in WinTalker, speaking the individual preview texts that you normally get when previewing their voices in the Mac's system preferences dialog.
WinTalker refers to a speak window that was developed in the 2000s, and purely consisted of the MacinTalk 3 voices. Keep in mind the speak window (and to an extent this recording) does not contain the high quality voices such as Alex or Bruce. Those didn't come later until possibly MacinTalk Pro.
WinTalker was developed by Mark Cecys, who sadly passed away in 2009 due to a rare form of cancer. He was the developer behind VocalWriter. He was also the developer behind an unreleased singing synthesizer for Microsoft's Whistler synthesis system (more info about this connection can be found in this blog post), and said samples are found in the Speech Synthesis Demonstrations/Microsoft/Whistler directory of this very archive.
This recording is straight from WinTalker's Speak to File function, with no extra post-processing or effects added to it.
The first DECtalk production in this folder, this was created using a build of DECtalk 5.1 from May 2008.
The production starts out as a reenactment of a sample of DECtalk 5.1 (May 2008) from Aksel Christoffersen, which is also in this archive.
Midway though, a new voice interrupts the introductions before Paul can sing the notes of a C Major chord.
DECtalk 5.1 features nine additional hidden voices that are in the files. While they aren't selectable in the speak windows for 5.1, they are still accessible using the traditional two letter commands to change voices, and they are mentioned in a few other files in the source code for 5.1 itself.
The new voices are:
This brings the total voice count to 18. Strangely though, both Ivan and Dennis use the same voice definitions for some strange reason, which results in them sounding nearly identical. As well as this, Jacky has a weird bug where her consonants are the only things that can be heard.
This is a recreation of the official Access Solutions DECtalk USB demo, created using the June 2006 build of DECtalk 4.63.
This uses the voice definitions from the DECtalk USB, taken from one of the various DECtalk source code archives, but in cases where voices started clipping/distorting, they were changed. It's likely that there's something else going on under the hood that affects the volume for each voice, so it's not entirely plug and play, but I tried the best I could here.