GoodBoyGoonsaLot SAM VOICE LAB
SAM engine · client-side
1982 FORMANT SYNTHESIS · MODERNIZED

Make weird voices.
Make them yours.

GoodBoyGoonsaLot wraps the original SAM formant engine in a modern voice-design workstation. Everything runs locally in your browser.

Ready to synthesize
01 · COMPOSE

Speech input

SAM's reciter is English-centric. English spelling and phonetic input are the most reliable.
0 / 4000 CtrlEnter preview
OUTPUT WAVEFORM—
Your generated waveform appears here
03 · PRESETS

Voice library

04 · CONTENT CLONE

Transcribe & Speak

WHISPER · OPTIONAL

Upload an MP3/WAV/M4A/etc. and transcribe its spoken content locally in the browser when the Whisper model is available. This copies words, not the speaker's identity, timbre or accent.

↥
Drop an audio file here or tap to choose · transcription runs client-side

For the most reliable recording format, use Record here; it creates a browser-native audio blob. Uploaded-file transcription accepts formats your browser can decode. Whisper is loaded only when you start transcription.

05 · HISTORY

Recent generations

ABOUT

GoodBoyGoonsaLot

This interface is a modern creative layer around the existing JavaScript SAM implementation. The formant synthesizer itself has deliberately not been replaced by a neural TTS engine.

SAM was originally published in 1982 by Don't Ask Software for the Commodore C64 and is a reverse-engineered speech system. This project preserves the included engine and its documentation/attribution.

Important: Content Clone means speech-to-text → SAM speech. It does not clone a person's voice identity.