Copy Assist — Speech to Text

Real-time speech-to-text transcription for the active slice. Uses whisper.cpp to decode received voice audio into a scrolling, color-coded transcript (green = high confidence, red = low). Includes model selection, compute device picker, and confidence-based text coloring.

Controls

Enable / Disable

Starts or stops the whisper.cpp speech-to-text engine on the active slice's audio.

Property	Value
Kind	Toggle button
Default	Disabled

Transcript

Scrolling, read-only transcript of decoded speech. Text is color-coded by whisper confidence: green (high), yellow (medium), red (low). Clear button empties the buffer.

Property	Value
Kind	Text field

Model tier

Whisper model size. Larger models are more accurate but slower and use more VRAM/RAM.

Property	Value
Kind	Combo box
Default	tiny
Valid range	tiny / base / small / medium

Compute device

Selects whether whisper runs on GPU (faster, needs VRAM) or CPU (slower, works everywhere).

Property	Value
Kind	Combo box
Default	GPU (CUDA/Metal)
Valid range	GPU / CPU

Backlog indicator

Seconds of received audio not yet transcribed. Colour escalates amber→red as backlog grows.

Property	Value
Kind	Indicator
Default	0.0s

Settings button

Opens the modeless Copy Assist settings dialog for model tier, compute device and engine configuration.

Property	Value
Kind	Push button

Clear

Clears the current transcript buffer.

Property	Value
Kind	Push button

Indicators

Engine status

Current state of the whisper.cpp speech-to-text engine.

State	Meaning
Idle	Engine is not running
Downloading model	Model is being downloaded
Loading model	Model is being loaded into memory
Listening	Engine is actively transcribing
Error	Engine encountered an error

Backlog

Seconds of untranscribed audio in the pipeline buffer.