Synthetic Sound Labs Scat-Talker

Image gallery

A vintage voice in a modern synthesizer

Synthetic Sound Labs’ Scat-Talker, Model 1420, gives a modular synthesizer the distinctive voice of a Votrax SC-01. Documented in December 2015 as a limited-edition module, it brings vintage speech hardware into a setting where its peculiar accent becomes a musical instrument.

The SC-01 synthesizes speech from phoneme codes using a model of the vocal tract. It is not playing recordings of a speaker. The Scat-Talker puts selection and pitch under panel and voltage control, inviting musicians to play the sounds themselves.

Patch a mouthful of sound

A control voltage chooses a speech sound; a gate or the illuminated panel button requests the selected phoneme. The display identifies the selection and gives an example word. Turning the selector alone does not immediately replace the sounding phoneme.

The module offers 63 sound-selection positions, including silence, plus a factory demonstration position. These are the module’s selections; the SC-01 itself uses a six-bit phoneme code. There is no text-entry interface or built-in spelling-to-speech converter. Words must be assembled by sequencing their constituent sounds.

Give the voice a melody

Pitch has its own panel control and two voltage inputs: a nominal 1V/octave input and a second input with adjustable modulation amount. The manufacturer describes the response as roughly exponential and cautions against expecting precise tuning or perfect stability.

That looseness is part of the instrument’s appeal. A gentle LFO can suggest vibrato; more extreme changes push the voice toward growls and cartoon chatter. A MIDI-to-CV converter can connect a computer-based sequencer to these inputs; MIDI is not a native connection on the module.

The timing that makes it musical

Most phonemes can sustain, while certain short, percussive sounds behave differently. The ACTIVE output indicates the chip’s built-in phoneme interval rather than simply copying the incoming gate. Audio can continue after that interval, and repeatedly requesting an unchanged selection does not produce the same new-phoneme activity.

In John L Rice’s demonstration, ACTIVE triggers an external envelope controlling a filter. A sequencer supplies phonemes while a keyboard supplies pitch, separating what the voice says from the notes it sings. Delay, reverb and phase processing heard in that demonstration come from external equipment.

From words to vocal percussion

Sound designer Ethan Persoff demonstrated another approach in 2025: a Rucci 16 Step CV Sequencer drives a Scat-Talker, while envelopes and a voltage-controlled amplifier trim speech sounds into musical events. Phonemes become percussion, bright lead sounds and slow drones.

Recognizable speech is only one destination. Repeating a small group of sounds, changing their pitch or shifting the sequence can create phrases that resemble a language without saying anything at all. The listener hears the shape of a voice, with rhythm and melody taking over the work of words.

A new life for the SC-01

The Scat-Talker belongs alongside the museum’s talking machines because it demonstrates a different use for their technology. The same kind of vocal-tract synthesis that gave early electronic voices their character can be explored as a playable sound source.

Its MU format is a single module unit wide and 5U tall. It is designed for a powered modular system, not as a self-contained speech accessory. A suitable power supply and other modules provide the surrounding instrument.

Specifications

Speech hardware

Votrax SC-01

Format

Single MU width; 5U tall

Phoneme selection CV

0 to +5 V

Gate input

Approximately 1.2 V logic threshold

Pitch CV

Nominal 1 V/octave; approximately 0 to +5 V; imperfect tracking

Modulation

Second pitch input with adjustable attenuation

Activity output

Logic high for the chip’s phoneme interval

Power connection

Synthesizers.com / DOTCOM six-pin Molex

Supply and current

+15 V at 73 mA; −15 V at 16 mA