Hardware synths that sing have mostly meant one thing for twenty years: load a vocal sample, pitch it with the keyboard, maybe run it through a vocoder. The Olydian Musical Instruments VC-01 Voicecrafter, revealed October 9 via a demo video and a Synth Anatomy writeup, skips that whole chain. It doesn't play back a voice. It assembles one, syllable by syllable, from a library of 380.
How does a machine build speech from scratch?
The VC-01 runs two voice engines side by side. One is formant synthesis, the technique that generates vowel-like tones algorithmically rather than recording them, the same family of math behind decades of talking synths and vintage speech chips. The other is consonant-vowel sampling, which plays back real recorded syllable fragments. Blend or crossfade the two and you get either a fully synthetic robot voice or something closer to chopped, re-sequenced human speech, selected from a touch panel large enough to browse all 380 syllables without digging through menus. Each engine carries its own programmable ADSR envelope and its own effects, so the formant half and the sample half don't have to sound like they came from the same throat.
Why should a house or techno producer care?
Sequenced, semi-intelligible vocal hooks are one of the oldest tricks in deep and tech house, from vocoder stabs to chopped diva samples cut into something closer to texture than lyric. Most producers build that effect in software, stitching together sample packs or running a vocal chain through a vocoder plugin. The VC-01 turns it into an instrument you play: punch in syllables on a 3-octave mechanical keyboard, sequence them with the built-in step sequencer (which Olydian says can hold thousands of steps with looping and overdubbing), and shape the result with two LFOs and a global FX section covering vibrato, reverb and three delay types. That's a tactile way to write a vocal hook or an eerie spoken-word break without clearing a sample or opening a DAW first.
The VC-01 is a large desktop unit featuring numerous controls, including both knobs and sliders, with a 3-octave mechanical Cherry-style keyboard and a pitch bend wheel.
When can you actually get one?
Nowhere yet. Olydian's own site lists price and release date as to-be-announced, with pre-orders expected to open sometime in 2027. This is Fakhoury's second instrument under the Olydian name after 2025's Photosynthesis, a solar-powered digital synth that made the same kind of left-field impression before it ever shipped. The VC-01 is still a prototype being shown off in a sound-demo video, not a shelf product, so the real test (how natural 380 syllables sound once they're actually chained together into a phrase) is still ahead.
Why it matters
Most hardware vocal tools are built around a fixed library of recorded phrases or a vocoder carrier signal. A machine that treats syllables like sequenced notes, with its own touch interface for picking them, is a genuinely new control surface for the oldest trick in dance music: making a voice say something it never actually said.
