SynthV Wiki
Advertisement
This article is still a work in progress

This article is slowly being improved for the benefit of all visitors. Please bear with us while improvements are being made. We apologise for the inconvenience this may cause in the meantime.
(What's Being worked on: Mixdown audio clips for Xingchen Minus and Yamine Renri R2; SynthV Engine versions (separate from but paired with SVS editor versions).)

Synthesizer V Studio is the second generation of the Synthesizer V vocal synthesis engine, developed by Dreamtonics Co., Ltd. The software was unveiled June 25, 2020 in a press release alongside voice databases Kotonoha Akane & Aoi and Saki.[1] AI support for Synthesizer V AI was released for Synthesizer V Studio as an update on December 25 alongside an update for Saki known as Saki AI. Voice databases that were not recorded with the AI method are known as "Standard voice databases", which are generally recorded at the Dreamtonics studio in Tokyo.[2][3]

Two versions were announced: Synthesizer V Studio Pro and Synthesizer V Studio Basic. Synthesizer V Studio Pro officially released July 10, 2020 on Beijing Photek S&T Development Co., Ltd.'s Taobao for Chinese users.[4] International users were able to purchase Synthesizer V Studio Pro starting July 15, on ANiCUTE,[5] while AH-Software Co. Ltd. later released it on July 30 for Japanese users. Synthesizer V Studio Basic released July 31, on Dreamtonics' Website as a free download.[6]

According to the press release, Synthesizer V Studio combines traditional sample-based voice synthesis with artificial intelligence and machine learning to create a hybrid engine.

History[]

In August 2019, Kanru Hua announced on Twitter the development of the second generation of Synthesizer V, known tentatively as Synthesizer V Release 2. On April 9, 2020 it was announced by Ddickky that the second generation of Synthesizer V would be released soon, and said a demo of Chiyu using the new engine would be coming soon.[7] The demo was released on April 11.[8]

Synthesizer V Studio Pro and Synthesizer V Studio Basic were formally announced on June 26 by AH-Software Co. Ltd. in a press release, as well as the voice databases Kotonoha Akane & Aoi and Saki.[9]

Animen's ANiCUTE store for international customers opened on July 12th and Synthesizer Studio Pro with voice databases Genbu & AiKO available for purchase on July 15.[10] Special discounts were made available for VIP members and purchasers of the 1st generation Synthesizer V editor.

Pre-orders for physical DVD versions of Synthesizer V Studio Pro with Kotonoha Akane & Aoi and Saki databases started on July 15[11] on AH-Software Co. Ltd.'s store and fully released on July 30.[12]

On August 2, Dreamtonics opened a beta-test application for VST and Audio Units versions of Synthesizer V Studio to anyone that purchased the software.[13]

Synthesizer V Studio was upgraded to version 1.1.0 on December 24 to support "Gen 1" AI voice databases.

Synthesizer V Studio was upgraded to version 1.2.0 on February 18, 2021 with AI voices now regarded as "Gen 2 AI voices".[14][15]

Synthesizer V Studio was upgraded to version 1.3.0 on June 18 with Synthesizer V AI being upgraded and now referred to as "Gen 3".[16][17]

On February 4, 2022, according to AH-Software, the Synthesizer V series had been selling much more than expected within a year ever since it became compatible with AI.[18]

Development of Neural Networks[]

Following the Synthesizer V Studio 1.2.0 Update on February 19, 2021, Kanru Hua announced a thread on his personal Twitter account about how the optimized neural network inference functions in the recent Synthesizer V updates.[19] The following day, Kanru Hua elaborated more on the subject. Synthesizer V Studio 1.2 uses JIT-compiled quantized sparse Matrix-vector multiplication (MVM) kernels.[20] In his own words "artificial neural network boils down to a bunch of really simple arithmetic operations, e.g. a + b * x1 + c * x2 + ... But when you (purposefully) compose millions of these together, they can be turned into really complicated machines."[21] He notes that in order to build a voice, he picks specialized values for the "a, b and c" that best represent the voice, and then plugs the values into millions of equations, These "a, b, c" values are called parameters. Linear algebra is used to help rather than writing each individual equation by hand to make matrices & vectors which are notations that aid with simple math in large batches. Many neural network models are composed of matrix-matrix multiplication. In the case of Synthesizer V, the bottleneck is matrix-vector multiplication, mainly used in a network that generates waveform samples which is known as the “neural vocoder”.[22] One of the challenges he notes is that not only having a large network to manage, the network needs to "run tens of thousands of times per second" to synthesize high quality audio in real time.[23] Due to this, modern CPUs are needed as they are able to run at several "gigacycles" per second ("GHz"). This is on a similar order of magnitude as the number of operations per second above however, he notes that the margin is very tight. He states that because of this, not all CPU cycles can do "useful work" and presents additional challenges.[24] The goal here is to make this MVM operation perform as fast as possible on modern CPU systems.[25]

The following day, he elaborated further into the usage of Sparse Matrix-Vector Multiplication (SpMVM) for Synthesizer V's neural networks.[26] Kanru Hua states that out of the millions of parameters, many are redundant and thrown out without hurting sound quality which results in what is called a "sparse matrix". [27][28] Some parameters that are truly important can't be thrown away and if too many are removed from too many of the parameters, eventually the quality will drop. "The synthesized voice will sound more and more like from a walkie-talkie until it completely degrades into noise."[29] The goal here is to remove the less contributing parameters carefully and remove as many as possible without hurting the quality. Typically over three-fourths of the parameters are thrown out if done properly.[30] When executing the sparse neural network, the program needs to skip the parameters that were removed. This skipping process adds a sometimes expensive overhead. This aids in boosting the speed up to four-times the initial speed.[31] "Going sparse is an effective way to compress a neural network. If done right, it can still speed up execution by a few times, although this would require highly optimized code for SpMVM."[32]

Over the course of the following three days, Kanru Hua posted three additional threads further elaborating on the intricises of developing the neural networks.[33][34][35] After making matrices sparse, integers are quantized to scale the values down before doing MVM to make sure the result will be in the range, it was noted that if the addition or multiplication programming goes out of range, they could be wrapped back to the lower end of the range resulting in an overflow and the synthesized voice could potentially sound akin to that of a mistuned radio or be complete noise.[36] He states that neural networks used in Synthesizer V AI come in many different sizes, some of which can be made sparse, some can not. The software is able to run on every x86 CPU since the Pentium 4 processor that was developed in 2004.[37]

Updates[]

Synthesizer V Studio 1.0.0 - date unknown (build 65536)[38] 
  • Original version delivered to demo producers and evaluators.
  • Synthesizer V Engine version unknown, but likely to be at least 2.0.0.
Synthesizer V Studio 1.0.3 Update - June 24, 2020 (build 65539) 
  • Version installed by the physical DVD-ROM release of Synthesizer V Studio Pro.
  • Synthesizer V Engine version 2.0.6 (as shown on the editor's About screen).[39]
  • Specific improvement(s) over earlier version(s) unknown.
Synthesizer V Studio 1.0.4 Update - July 9, 2020 (build 65540) 
  • Version obtained by digital download upon first public release (both Basic and Pro editors); first public update issued to users who bought physical editions of Pro.
  • Still based on Synthesizer V Engine 2.0.6.[40]

Bug Fixes

  • Fixed a stability issue noted by producer Amano-kei in 1.0.3, where rendering lagged if too many unsaved edits were present (thus requiring frequent saves).[41][42]
Synthesizer V Studio 1.0.5 Update - August 14, 2020 (build 65541)[43] 
  • Synthesizer V Engine updated to 2.0.7.[44]

Enhancements

  • Side panels: select the value when double clicking a slider
  • Linguistics: allow override of phonemes on notes with syllable progression symbol ("+")
  • About window: add system information for bug reports

Bug Fixes

  • Audio: failure to select output devices without recording capability
  • Live Render: crashes when deleting a track during playback
  • Synthesizer V engine: crashes on rare phoneme combinations
  • File: tracks with same name are not loaded
  • Parameter panel: erase mode sets voicing and vibrato envelope parameters to zero, instead of the default value
  • Scripting: errors in return values of Automation.getPoints()
  • Scripting: crashes when launching a dialog from a callback function
Synthesizer V Studio 1.0.6 Update - September 4, 2020[45] 
  • Synthesizer V Engine upgraded to version 2.0.8.

New Features

  • VST3 plugin version for Synthesizer V Studio Pro (Windows / macOS).
  • AU plugin version for Synthesizer V Studio Pro (macOS).

Enhancements

  • Piano roll: extend the scroll range to be the same as arrangement view.
  • Piano roll: improve the snapping behavior when dragging notes.
  • Editing: add support for zoom gesture on touchpads.
  • Editing: use Mac delete key for deleting the selection (macOS).
  • Other: display file name on the title bar.

Bug Fixes

  • Synthesizer V engine: audio artifacts and potential crashes at extremely high pitches.
  • Piano roll: notes created by double clicks are not of correct length (macOS).
  • Piano roll: selection rectangle shows up when creating a note beyond the work area.
  • Playback: spacebar keybinding for play/pause cannot be overridden.
  • Scripting: keybindings for scripts get reset after restart.
Synthesizer V Studio 1.0.8 Update - September 5, 2020 (build 65544)[46] 
  • Next-day update to address the issue that some users could not use shortcut keys in 1.0.6; otherwise "there seems to be no new features added."
  • Still based on Synthesizer V Engine 2.0.8.[47]
Synthesizer V Studio 1.1.0 Update - December 24, 2020 (build 65792)[48] 
  • Synthesizer V Engine updated to version 2.1.2.[49]

New features

  • Support for Al voice databases.
  • Voice panel add two render modes 'Prefer Speed" and "Prefer Quality" (only for Al voice databases).
  • Modify menu: add second generation Auto Pitch Tuning. Compared to the first generation, pitch curves generated are more natural. The Pro edition allows control over the degree of variation and randomization.
  • Parameter panel: add "Tone Shift parameter (only for Al voice database, only for Pro edition) which allows changing timbre but fixing the pitch. Similar to "Expression Group" for standard voice databases.
  • Note properties: add phoneme "Strength" property (only for Al voice database, only for Pro edition).
  • Piano roll: add note detuning; pitch adjustments in 25 cents (¼ semitone) units is possible by dragging notes while holding Ctrl + Shift (only for Pro edition).
  • Plano roll: add "Split Note/Group" feature in the right-click menu.
  • Piano roll: add resume mode. If on (by default), playback will resume from the paused position. If off, playback will resume from marker/start position.
  • Piano roll: add loop mode and loop markers.
  • Transport menu: add "Loop Over Selected Notes" feature.
  • Misc: add Neural Networks Performance Tuner (only for Pro edition) which optimizes Al voice rendering for the clent machine.

Enhancements

  • Live Rendering System: reduced latency and overhead of LRS.
  • Piano roll: add special lyrics "br" and "cl" (only for Al voice databases).
  • Piano roll: add link button to simultaneously switch between arrow/pencil tools and notes/pitch editing modes.
  • Note properties: add slider reset feature - double clicking on the slider for phoneme duration or strength will reset it to 100%.
  • Settings: add "Show animated rendering status" option.
  • Scripting: add following methods.
    • MainEditorView.setCurrentTrack()
    • MainEditorView.setCurrentGroup()
    • TrackInnerSelectionState.getSelectedPoints()
    • TrackInnerSelectionState.select Points()
    • TrackInnerSelectionState.unselectPoints()
Synthesizer V Studio 1.1.1 Update - December 26, 2020 (build 65793)[50] 
  • Still based on Synthesizer V Engine 2.1.2.
  • Fixed issues reported with AI databases in the previous version for 2nd, 3rd gen Intel i3/i5/i7 processors and AMD processors.
Synthesizer V Studio 1.2.0 Update - February 18, 2021[51][52] 

40% - 75% Faster Rendering

  • Gen 2 Al voices achieve significant speed up with quantized sparse linear algebra kernels.
  • This is a combination of techniques that compresses a large neural network into a smaller version without losing quality.

More Responsive

  • 120% File Size Reduction
    • The entire voice installer fits into 10 MB.
    • Just a few seconds to download.

Fewer "Surprises"

  • Neural network based waveform generation techniques often suffer from random glitches. The chance of occurrence was 1 in 400 seconds.
  • The stablized WaveRNN vocoder reduces these glitches down to a nearly undetectable rate.

Robust Pronunciation

  • Gen 1 AI voices tend to mispronounce rare phonemes, in particular the Japanese yoōn's
  • Gen 2 AI voices are built with care for rare cases.
    • No more mispronunciations.

Cleaner Sound

  • Reduced noise in high frequency range.
  • More spectral details preserved.
Synthesizer V Studio 1.2.1 Update - February 19, 2021[53] 

New features

  • Fixes a bug causing crashes and file corruption during voice updates.
  • Live Render: add power management settings to throttle the synthesis engine when rendering out-of-screen contents.
  • Plugin: allow Synthesizer V Studio projects to be saved inside the DAW's project file (enabled by default).
  • Scripting: the Pro version now comes with scripts such as “Remove Short Silences”.

Enhancements

  • Synthesis engine: significant optimization on Al voice rendering.
  • Synthesis engine: support for stabilized voice models.
  • Piano Roll: add "Insert Lyrics..." to the right-click menu.

Bug Fixes

  • Playback: looping stops when the playhead goes beyond the project's duration.
  • Playback: plugin ceases to follow host's playback status once going beyond the project's duration.
  • Update: file corruption when updating voices.
  • Scripting: Project.getNoteGroup fails when there are more groups than tracks.
Synthesizer V Studio 1.2.2 Update - March 25, 2021 (build 66050)[54][55] 
  • Synthesizer V Engine version 2.2.1.[56]
  • 1.2.2 update brings bug fixes, various minor changes and stability improvements.

New Features

  • Arrangement: add a "Duplicate Track" command to the right-click menu on track items.
  • Editing: add a "Dissociate Group" command that takes out the contents from a note group and form a new, independent note group.

Enhancements

  • Arrangement: update the display range of note groups during edits.
  • Piano roll: display quarter beat grid even when snapping is disabled.

Bug Fixes

  • Synthesizer V Engine: occasional bursts of noise in Al rendered voices.
  • Linguistics: broken support for "brl + number" type special lyrics for long breaths.
  • Plugin: crashes when loading projects in some certain DAWS.
  • Plugin: Input method editor is disabled in "Input Lyrics Dialog" in Windows VSTi version; multi-line input doesn't work under MacOS.
Synthesizer V Studio 1.2.4 Update - June 3, 2021 (build 66052)[57][58] 
  • Synthesizer V Engine updated to 2.2.3.[59]
  • A feature update to allow for Tsurumaki Maki Standard & AI voices to be used in Synthesizer V Studio, however, a memory leak was found by users utilizing the Lite voices and an update to 1.2.5 was released immediately the following day.
Synthesizer V Studio 1.2.5 Update - June 4, 2021[60][61] 
  • After reports of software crashes when selecting the output folder from the Rendering panel of Synthesizer V Studio, a fix was applied from version 1.2.4. Engine version still 2.2.3.
Synthesizer V Studio 1.2.6 Update - June 4, 2021  
  • A further update from 1.2.5 released the same day applying a fix to further crash reports. Engine still version 2.2.3.
Synthesizer V Studio 1.3.0 Update - June 18, 2021 (build 66304)[62][63][64] 
  • Synthesizer V Engine still shown as version 2.2.3.[65]
  • "Synthesizer V Studio 1.3.0, available from June 18, brings significant naturalness improvement to AI voices together with a collection of features requested by our users."

New Features

  • Synthesizer V Engine: support for third generation Synthesizer V AI voices (also requires updating the voices).
  • Voice panel: add “Tuning” properties for pitch shifting the entire group/track (only for Pro edition).
  • UI: add a bottom status bar for displaying usage tips.

Enhancements

  • Editing: add an option to paste at the playhead.
  • Editing: add a few modifier keys related to dragging behaviors:
    • Hold Ctrl/Command + Alt while dragging to temporarily disable snapping.
    • Hold Ctrl/Command + Shift while dragging to slow down the movement of control points.
  • Piano Roll: improve the display of control points while dragging.
  • Piano Roll: add an adaptive snapping mode to adjust the grid based on the zoom level.
  • Piano Roll: add a snapping mode selector to piano roll’s toolbar.
  • License and Updates panel: improve the layout of the product list.

Bug Fixes

  • Editing: crashes when notes or groups are pasted before the first beat.
  • Editing: crashes when an entire group is moved before the first beat.
Synthesizer V Studio 1.4.0 Update - November 18, 2021 (build 66560)[66][67][68] 
  • Synthesizer V Engine updated to version 2.3.0.[69]
  • "This update brings support for the 4th gen AI voice databases, a new Auto Pitch Tuning command allowing customization, a command for auto-splitting notes based on the melody, improved phoneme conversion for American English and other feature and sound quality enhancements."

New Features

  • Auto-Process: regroup auto pitch tuning related features under “Auto-Process” menu.
  • Auto-Process: add support for 4th generation pitch models.
  • Auto-Process: add a customizable version of Auto Pitch Tuning independent of the voice database. (only for Pro edition)
  • Auto-Process: add “Ornament Selected Notes” command that automatically splits the notes based on the melody context. (only for Pro edition)
  • Voice panel: add “Use relaxed consonants” option for phoneme conversion rules that better suit American English.
  • Lyrics Processing: new lyrics shorthand that inserts a glottal stop symbol (“cl”) for notes starting with a single quote.

Enhancements

  • Synthesizer V Engine: render speed and sound quality improvement for the latest AI voices.
  • Linguistics: improve the handling of English lyrics containing single quotes (e.g. dreamin’, I’m, didn’t).

Bug Fixes

  • Playback: muted tracks are audible during the first few hundred milliseconds of playback.
  • Plugin: mute/solo/gain status of instrumental tracks are ignored during mixdown.
  • Plugin: timing mismatch during playback in Cubase.
  • Other: rare crashes when opening/close a side panel.
Synthesizer V Studio 1.4.1 Update - December 16, 2021 (build 66561)[70][71] 
  • Synthesizer V Engine updated to 2.3.1.[72]
  • "This update includes stability improvements to the synthesis engine and numerous bug fixes."

Specification Changes

  • Arrangement: allow groups to be dragged beyond the start of the timeline.

Bug Fixes

  • Synthesizer V Engine: small quality degradation on 4th generation voices under Prefer Quality mode.
  • Auto-Process: crashes at the end of Auto Pitch Tuning.
  • Live render: crashes during rendering and playback.
  • License and Updates panel: crashes when deactivating a voice.
  • License and Updates panel: crashes when closing the panel with products pending refreshes.
  • Piano roll: the first click after switching to freehand Direct Pitch Editing mode creates an unexpected note.
  • Plugin: playback gets stuck when triggered from the plugin.
  • Other: the standalone editor displays as “Synthesizer V Studio Plugin” under macOS.
  • Other: unable to save settings when a corrupted settings file is present.
Synthesizer V Studio 1.5.0 Update - December 23, 2021 (build 66816)[73][74] 
  • Synthesizer V Engine upgraded to version 2.3.2.[75]
  • "Version 1.5.0 brings another round of comprehensive upgrades to AI based Synthesizer V voices, marking a major milestone towards bringing realistic synthesized voices to serious music production workflows."

New Features

  • Synthesizer V Engine: cross-lingual singing synthesis for AI voices (only for Pro edition), configurable via “Sing in the following language” in Voice panel.
  • Piano roll: automatically shrink the neighboring note when enlarging a note; togglable via “overlap-less note resizing” in Settings panel.
  • Other: add crash recovery.

Specification Changes

  • Auto-Process: add a default shortcut (Alt + R) for re-running Auto Pitch Tuning with a different random seed.
  • Editing: menu item “Reset Pitch Properties” is renamed to “Reset Pitch”, which will reset both the pitch properties for a note and the pitch deviation parameter under the note.
  • Piano roll: add reset commands and “Select Parameter for Notes” command to the context menu on selected notes.

Enhancements

  • Synthesizer V Engine: add support for high rendering speed and high spectrogram resolution models for AI voices.
  • Auto-Process: improved expressiveness for Auto Pitch Tuning.
  • Playback: add an animated highlight effect to notify users when a track is played without a voice database loaded.
  • Rendering: improved resampling quality when exporting to a different rate from the engine’s internal sampling rate.

Bug Fixes

  • Piano roll: crashes when dragging notes selected in some certain order.
  • File: the order of tracks gets reset when loading a project with many tracks.
  • Plugin: playback gets stuck when triggered from the plugin on macOS.
Synthesizer V Studio 1.5.1 Update (macOS only) - February 22, 2022 (build 66817)[76][77] 
  • Dual x86-64/ARM64 (Universal Binary Format) port of version 1.5.0 (as such, Synthesizer V Engine version is still 2.3.2).[78]
    • Prior to 1.5.1, to run any version of SynthV Studio on a Mac computer with an Apple Silicon M1 or higher CPU (as opposed to an Intel CPU), Rosetta 2 (a compatibility emulation layer that transcodes Intel x86-64 machine code to Apple-specific ARM64 instructions) was required.
  • "Synthesizer V Studio now runs natively on Apple Silicon! Get the version 1.5.1 update to get a 300% rendering speed boost on M1 series chips. For Pro version users, native Apple Silicon support also extends to VST3 and AU plugins."
  • "Starting from 1.5.1, the macOS versions of Synthesizer V Studio are available in the universal binary format, which includes native Apple Silicon support.
    On processors such as Apple M1, version 1.5.1 archives a staggering rendering speed of 300% against the previous versions which run through the Rosetta 2 compatibility layer on the same hardware. To put this into some context, the demo song “Stir Crazy” for Synthesizer V AI Kevin can be rendered in 10 seconds.
    Synthesizer V Studio Pro 1.5.1 also works as VST3/Audio Unit plugins on DAWs with universal binary plugin support."

Bug fixes

  • File: .svp file association does not work on macOS.
  • Plugin: some key presses on the plugin got leaked into the DAW on macOS.
Synthesizer V Studio 1.5.2 Update (macOS only) - February 25, 2022[79] 
  • Fixes crashes when using Standard voice databases in some Apple Silicon based environments.
Synthesizer V Studio 1.6.0 Update - March 25, 2022 (build 67072)[80][81] 
  • Synthesizer V Engine upgraded to version 2.4.0.
  • With the 1.6.0 update, Synthesizer V Studio Pro now supports switching between a few predefined vocal styles for AI voice databases.
  • Vocal Mode: Creators can now use the “Vocal Mode” section in the Voice panel to select among or mix a handful of presets such as “chest”, “open”, “soft”, “clear”, thereby customizing out a richer and more versatile sound tailored for their work.
  • Instant Mode: Another important addition to the singing editor is the “Instant Mode”. While the Auto Pitch Tuning feature is bringing great naturalness improvement at the merely cost of a single menu command, we further simplified the workflow by letting the editor execute Auto Pitch Tuning as soon as the user modifies any note or lyrics. Behaviors of the pitch generation AI can even be customized for each note, for example increasing/reducing the pitch dynamics only for some selected notes.
  • Voice Updates: Due to the varying styles and specifications of the voice databases, the supported Vocal Mode options also vary. The first batch of AI voice databases with Vocal Mode support are Dreamtonics’ original voices and Stardust (by Quadimension). We are working with our licensed partners to bring Vocal Mode feature to more voices in the future.

New Features

  • Synthesizer V Engine: Vocal Mode properties for AI voices (only for Pro edition), configurable under “Vocal Mode” section in Voice Panel.
  • Auto-Process: add Instant Mode under which Auto Pitch Tuning is executed immediately after modifying notes and lyrics.
  • Note properties panel: add note-level customization for Auto Pitch Tuning parameters.
  • Settings: add a slider for controlling the GUI scaling factor.

Enhancements

  • Auto-Process: improved expressiveness further for Auto Pitch Tuning.
  • Arrangement: wider range for horizontal scaling and improved display of measure numbers.

Bug Fixes

  • File: notes with zero duration are created when loading some MIDI files. Editing tracks containing these problematic notes may lead to crashes.
  • Parameter panel: fix crashes when editing control points selected with “Select Parameters for Notes”.
  • Editing: fix crashes when executing certain menu commands (e.g. “Snap to Grid”) or scripts while dragging notes or groups.
Synthesizer V Studio 1.6.1 Update - April 7, 2022 (build 67073)[82][83] 
  • Synthesizer V Engine version updated to 2.4.1.
  • Synthesizer V Studio 1.6.1 update has been released.
  • This update includes various fixes for crashes and compatibility issues.

Bug Fixes

  • Auto-Process: crashes when loading a standard voice database without auto pitch tuning support under Instant Mode (Basic edition only).
  • Synthesizer V Engine: random audio glitches when using AI voices on Intel Macs.
  • File: Instant Mode cannot be turned off by default when importing MIDI files.
  • File: audio files imported by drag-and-drop are not indexed by relative paths.
  • Plugin: broken IME support in the Windows VSTi version.

Requirements[]

  • OS: Windows 8.1 or later (32/64 bit), macOS 10.11 (El Capitan) or later, Linux Ubuntu 16.04 or later (64 bit).[84][85][86]
    • Dreamtonics' Japanese and English pages disagree on the minimum Ubuntu requirements between Basic and Pro editions, with the English page specifying 16.04+ for Basic and 18.04+ for Pro and the Japanese page swapping both values (18.04+ Basic, 16.04+ Pro).[84][87] AH-Software maintained 16.4+ as minimum Ubuntu spec for both Basic and Pro until around October 2021, when they increased it to 18.04+ (64-bit) for both Basic and Pro.[88]
    • In November 2021, with the release of Windows 11 and the retirement of 32-bit support, AHS reworded the OS spec to: Windows 11/10/8.1 or later (64-bit); Mac OS X: 10.11 or later; Ubuntu 18.04 or later (64-bit).[89]
  • CPU: (x86-64) Intel Core i3 or higher, or equivalent AMD processor;[85] (ARM64) Apple Silicon M1 series.[76]
    • Dreamtonics' page specifies instead: Intel Core i5 or higher, or equivalent AMD processor.[90] AHS did not match this specification online until around October 2021.[88]
    • For high-quality editing of AI audio on x86-64 platforms (also known as AMD64, Intel EM64T, or simply x64): Intel Core series 4th generation i5 (i5-4xxx) or higher is recommended, or AMD Athlon X4 845 or higher, including Ryzen series.[91]
  • RAM: 2GB or more.
  • HDD: 1GB or more (for one voice database installation).
    • Additional space will be required according to the number of voice databases to be installed.
  • Display resolution: 1280x800 or higher resolution.
  • Others: DVD-ROM drive (physical version only), audio device, internet connection.

Differences between Basic and Pro[]

Basic Pro
Project limitations Limited to 3 tracks Unlimited tracks
Simultaneous rendering thread Limited to 2 cores Unlimited rendering thread
Additional features
  • Auto pitch tuning (default configurations only)[92]
  • Auto pitch tuning (with customizeable configurations)
  • Aspiration output
  • Lua/JavaScript plugin
  • Alternative phonemes
  • Additional parameters for AI databases
VST/AU plugin Not available Available

Releases[]

Full Versions[]

  • Full versions are the paid versions of the voice databases (with the exception of Yamine Renri, which is free to register and download).*
  • These are known as the Synthesizer V Standard voice databases. Synthesizer V AI voice databases are not included in this list.

Lite Versions[]

  • Lite versions are free versions of the voice databases and are available on Synthesizer V Studio Basic and Pro.
  • These are generally monopitch (1 pitch) voice databases.
  • General guidelines include mentioning "Lite" as part of an uploads' credit in the title and description, as well as disallowing commercial use. Further guidelines may depend on each voice database and it is encouraged that they are read carefully.
  • These are considered the Synthesizer V Studio Standard voice databases. The Synthesizer V AI lite editions are not listed.

Note: Not every voice database has a Lite version.

Starter Packs[]

  • Full versions that can be acquired together with the Synthesizer V Studio Pro at a discount.
  • ANiCUTE Bundle Packs, Dreamtonics Bundles, and Taobao Sets consist of the voice databases themselves bundled with the SVS Pro editor in the same digital download; they do not exist as physical bundles.
    • The 4-in-1 Bundle Pack featuring Chiyu, Haiyi, Shian, and Cangqiong is the only ANiCUTE Bundle Pack that does not come with the SVS Pro editor. Instead, it features only the four voice databases.
  • AH-Software starters (marked with * below) are just standalone voice databases that can be freely downloaded from the AHS Store after redeeming the coupon enclosed within the (physical-only) Synthesizer V Studio Pro Starter Pack. Only one of them can be downloaded per Starter Pack coupon.
  • These are Synthesizer V Standard voice databases. Synthesizer V AI starters are not listed here.

Note: Not every voice database has a Starter Pack.

Announced Vocals[]

Discontinued Bundle Packs[]

Differences: Lite vs. Full[]

Differences between "Lite" vocals and "Full" vocals.

Lite Full
Expression Group One pitch group only 3 pitches or more
Auto pitch tuning None Depending on the Voice Database
Commercial usage Not permitted[93] Permitted
Attribution Mentioning "This work uses the Lite version of the Voice Database" is required Not required

Known Problems[]

  • Synthesizer V Studio 1.1.0
    • Dreamtonics noted that some users confirmed crash reports related to AI voices. This bug affects 2nd, 3rd gen Intel i3/i5/i7 processors and AMD processors from Jaguar to Steamroller series. Dreamtonics has recommended upgrading to Synthesizer V Studio 1.1.1[94]
  • Synthesizer V Studio 1.2.0
    • Dreamtonics noted that some users are experiencing crashes while updating Saki AI to version 104, they have recommended directly downloading & installing the voice from their website. They are currently looking into this issue and will release a fix as soon as possible.[95]

Additional installation notes[]

Additional notes[]

New features[]

New interface that focuses on information concentration and readability. Singing edits render instantly.

Multi-functional piano roll offering near real-time feedback.

Parameter curve tuning function with freehand parameter curve editing, smooth parameter curve tuning that uses a new parameter panel with control points.

Faster rendering speed and improved functionality over Synthesizer V Studio Basic version. Includes a singing voice database and allows access to the "Web Synthesizer V" version. "Breath component separation output" rendering option, which allows the singing voice model to separate the breath component from the singing voice. [96]

Examples of usage[]

Demonstrations[]

Demos of Kotonoha Akane & Aoi's voice libraries can be found on their promotional pages.[97]

Gallery[]

Packages[]

Download Icons[]

Logos[]

Media Gallery[]

References[]

  1. https://twitter.com/ahsoft/status/1276351253073629185
  2. https://twitter.com/OfficialVolor/status/1392293297847037954
  3. https://www.youtube.com/watch?v=ET6Kdw15L_8
  4. https://www.weibo.com/5056093502/J8zzR6Xk9?type=comment#_rnd1597881463992
  5. https://twitter.com/dreamtonics_en/status/1283386080251830272
  6. https://twitter.com/dreamtonics_en/status/1277908151807819776
  7. https://www.weibo.com/5056093502/ICxlpBPw4
  8. https://www.bilibili.com/video/BV1ba4y1x7pg
  9. https://www.ah-soft.com/press/synth-v/
  10. https://twitter.com/dreamtonics_en/status/1283386080251830272
  11. https://twitter.com/ahsoft/status/1283571514344747008
  12. https://twitter.com/ahsoft/status/1288639895339036673
  13. https://twitter.com/dreamtonics_en/status/1289851729668747271
  14. https://twitter.com/dreamtonics_en/status/1362598004100800519
  15. https://www.youtube.com/watch?v=D_cXLpO0qIw
  16. https://twitter.com/dreamtonics_en/status/1362598004100800519
  17. https://www.youtube.com/watch?v=D_cXLpO0qIw
  18. https://www.ah-soft.com/press/synth-v/20220204.html
  19. https://twitter.com/khuasw/status/1362799523156746240
  20. https://twitter.com/khuasw/status/1363069116467138564
  21. https://twitter.com/khuasw/status/1363073616821116931
  22. https://twitter.com/khuasw/status/1363073620445192195
  23. https://twitter.com/khuasw/status/1363073621741084680
  24. https://twitter.com/khuasw/status/1363073622860992520
  25. https://twitter.com/khuasw/status/1363073624010235905
  26. https://twitter.com/khuasw/status/1363431631759925259
  27. https://twitter.com/khuasw/status/1363431631759925259
  28. https://twitter.com/khuasw/status/1363431633404108803
  29. https://twitter.com/khuasw/status/1363431634670817287
  30. https://twitter.com/khuasw/status/1363431636012986371
  31. https://twitter.com/khuasw/status/1363431637506170882
  32. https://twitter.com/khuasw/status/1363431640085635075
  33. https://twitter.com/khuasw/status/1363828518686138375
  34. https://twitter.com/khuasw/status/1364164654231031815
  35. https://twitter.com/khuasw/status/1364774324599615494
  36. https://twitter.com/khuasw/status/1364164660132372481
  37. https://twitter.com/khuasw/status/1364774332874985474
  38. Build version number 65536 is only known from a screenshot embedded in a DTM Station article describing Synthesizer V Studio updates 1.0.6 and 1.0.8, linked later below.
  39. About screen for SynthV Studio Pro version shipped in the physical DVD
  40. About screen for original SynthV Studio Basic version downloaded by users at launch date
  41. https://amanokei.hatenablog.com/entry/2020/07/10/002728 (in Japanese)
  42. https://vocaverse.network/threads/synthesizer-v-discussion-thread.202/post-27312 (partial translation of the above by xuu)
  43. https://twitter.com/dreamtonics_en/status/1294260623623086080
  44. Screenshot of SVS Pro 1.0.5 showing build number and engine version
  45. https://twitter.com/dreamtonics_en/status/1301814760883851265
  46. https://www.dtmstation.com/archives/31496.html
  47. Screenshot of SVS Pro 1.0.8 showing build number and engine version
  48. https://twitter.com/dreamtonics_en/status/1342304090861182977
  49. Screenshot of SVS Pro 1.1.0 showing build number and engine version
  50. https://twitter.com/dreamtonics_en/status/1342732089712492545
  51. https://twitter.com/dreamtonics_en/status/1362598004100800519
  52. https://youtu.be/D_cXLpO0qIw
  53. https://twitter.com/dreamtonics_en/status/1362709004086964226
  54. https://twitter.com/dreamtonics_en/status/1362598004100800519
  55. https://www.youtube.com/watch?v=D_cXLpO0qIw
  56. Screenshot of SVS Pro 1.2.2 showing build number and engine version
  57. https://twitter.com/ahsoft/status/1400349973749768195
  58. The installer files for Synthesizer V Studio Basic 1.2.4 included in the physical Tsurumaki Maki AI Complete and English AI Complete DVD-ROMs were date-stamped to May 16, 2021, meaning the update was finalized more than 2 weeks before it was announced on Twitter on June 3.
  59. Screenshot of SVS Basic 1.2.4 showing build number and engine version
  60. https://twitter.com/dreamtonics_en/status/1400674631279812609
  61. https://www.youtube.com/watch?v=D_cXLpO0qIw
  62. https://dreamtonics.com/en/synthesizer-v-studio-1-3-0-update/
  63. https://twitter.com/dreamtonics_en/status/1405721939235131393
  64. https://www.youtube.com/watch?v=ir3eb9lwKp8
  65. Screenshot of SVS Pro 1.3.0 showing build number and engine version
  66. https://dreamtonics.com/en/synthesizer-v-studio-1-4-0-update/
  67. https://twitter.com/dreamtonics_en/status/1461288124676407302
  68. https://twitter.com/dreamtonics_en/status/1461288126010179589
  69. Screenshot of SVS Pro 1.4.0c showing build number and engine version
  70. https://dreamtonics.com/en/synthesizer-v-studio-1-4-1-update/
  71. https://twitter.com/dreamtonics_en/status/1471404820850700289
  72. Screenshot of SVS Pro 1.4.1c showing build number and engine version
  73. https://dreamtonics.com/en/cross-lingual-support-for-synthesizer-v-ai/
  74. https://twitter.com/dreamtonics_en/status/1473941453810577410
  75. Screenshot of SVS Pro 1.5.0's About screen showing engine version
  76. 76.0 76.1 https://twitter.com/dreamtonics_en/status/1496047565959217154
  77. https://dreamtonics.com/en/native-apple-silicon-support-for-synthesizer-v-studio/
  78. Screenshots of SVS Pro 1.5.1's About screen (showing engine version) and Update pane (showing build version number)
  79. https://twitter.com/dreamtonics_en/status/1497122610211332102
  80. https://dreamtonics.com/en/synthesizer-v-studio-1-6-0-update/
  81. https://twitter.com/dreamtonics_en/status/1507270199510470663
  82. https://dreamtonics.com/en/synthesizer-v-studio-1-6-1-update/
  83. https://twitter.com/dreamtonics_en/status/1511977292746358784
  84. 84.0 84.1 Dreamtonics Co., Ltd.'s Synthesizer V Studio page (in English), archived June 30, 2020
  85. 85.0 85.1 AH-Software Co. Ltd.'s Synthesizer V Studio physical box, as well as its specifications page (as archived June 26, 2020). Among other things, it did not specify 64-bit requirement for Linux Ubuntu (in addition to misprinting 16.04 as 16.4) and maintained i3 as minimum Intel Core CPU specification.
  86. AH-Software Co. Ltd.'s Synthesizer V Studio specifications page, archived November 27, 2020 (moved to new URL, same specifications otherwise)
  87. Dreamtonics Co., Ltd.'s Synthesizer V Studio page (in Japanese), archived August 5, 2020
  88. 88.0 88.1 AH-Software Co. Ltd.'s Synthesizer V Studio specifications page, archived November 2, 2021
  89. AH-Software Co. Ltd.'s Synthesizer V Studio specifications page, archived November 5, 2021
  90. Dreamtonics Co., Ltd.'s Synthesizer V Studio page, archived September 21, 2020
  91. AH-Software Co. Ltd.'s Synthesizer V Saki specifications page, archived May 13, 2021
  92. Some voice databases don't have auto pitch informations, therefore, this feature can't be used for said voice databases.
  93. Even though that's what's written on the download website, every voice database has different commercial license agreements.
  94. https://twitter.com/dreamtonics_en/status/1342732089712492545
  95. https://twitter.com/CeVIO_st/status/1355085295163596807
  96. https://www.ah-soft.com/synth-v/index.html#feature
  97. https://www.ah-soft.com/synth-v/kotonoha/#top
  98. https://www.ah-soft.net/shopdetail/000000001204/ct120/page1/order/
  99. https://www.ah-soft.net/shopdetail/000000001204/ct120/page1/order/
  100. https://www.ah-soft.net/shopdetail/000000001202/ct120/page1/order/
  101. https://www.ah-soft.net/shopdetail/000000001206/ct120/page1/order/
  102. https://www.ah-soft.net/shopdetail/000000001206/ct120/page1/order/
  103. https://www.ah-soft.net/shopdetail/000000001203/ct120/page1/order/
  104. https://www.ah-soft.net/shopdetail/000000001294/ct120/page1/order/
  105. https://www.ah-soft.net/shopdetail/000000001295/ct120/page1/order/
  106. https://www.ah-soft.net/shopdetail/000000001296/ct120/page1/order/

External links[]

Navigation[]

Advertisement