DirectorKit Research

Experiments and research progression

A small public-facing index for prototypes, listening tests and demos that show how the speech stack is improving over time.

Current demos in apps/experiments

Emotion Control TTS Demo

CLSP ControlNet Echo-TTS

Interactive Echo-TTS editing demo for steering generated voice style and emotion with CLSP ControlNet conditioning.

Open emotion-control TTS demo
New Voice Identity Shifting Demo

VoiceID Vector Shifter

Upload a clip, shift emotions of the clip while retaining the identity by adjusting the identity vector in real-time.

Open voice identity shifting demo
New Video Page

Early Translation & Localization Testing

Video demo capture of early testing on localization and translations workflows using Voice-DAW app.

Open video demo
Word-less Emotion Transform Demo

VoiceTransformDiT Demo

Browser demo for text prompt based emotion transformations focused on non-word based human sounds.

Open demo
Interactive Audio

CLSP Sphere Sorting Demo

Audits nearest-neighbor sample selection along CLSP prompt axes such as calm/nervous, shouting/whispering, and child/old person.

Open CLSP demo
Research Harness

Timestretch Algorithm Evaluation

Compares waveform stretch, Ruchey-lane stretch, and SDEdit-assisted stretch strategies for vowel-biased timing edits.

Open evaluation demo
Recording Tool

Voice Emotion Dataset Recording

Browser control surface for recording actors and dataset capture workflows used by the voice data pipeline.

Open app
New Listening Page

Ruchey Audio Quality Progression

Compare source clips, old checkpoints and passes, and newer model generations in one repeatable listening-test layout.

Open progression page