Advanced Settings

Advanced Settings

The Advanced page covers history retention, whether runs are saved, model memory behavior, long-term memory, and spoken feedback.

📸 Screenshot needed: The Advanced page in full, showing history retention mode, the Save to history toggle, the model auto-unload idle timeout, the memory enable toggle, and the spoken-feedback controls.

History retention

You choose how long transcription history is kept:

Mode	Behavior
Duration	Keeps history for a set period. Defaults to 90 days.
Forever	Keeps history indefinitely.
Until app closes	Discards history when the app exits.

📸 Screenshot needed: The history retention mode selector with Duration (90 days), Forever, and Until app closes options visible.

Save to history

Use the Save to history toggle to skip storing runs you do not want kept.

See History for searching and managing stored transcriptions.

Model auto-unload

TypeWhisper can unload the active model after a configurable idle timeout to free resources. Setting the timeout to 0 disables auto-unload, keeping the model loaded.

📸 Screenshot needed: The model auto-unload idle-timeout field, including the note that 0 disables auto-unload.

Memory

The Memory enable toggle turns on optional long-term memory. It is gated: it only becomes available when you have both a memory storage plugin and an available LLM provider.

📸 Screenshot needed: The memory enable toggle in both its gated (disabled) and enabled states.

When enabled, eligible transcriptions are sent to the LLM to extract durable facts that future prompt actions can recall as context. See Memory for details.

Spoken feedback

You can have TypeWhisper speak feedback aloud.

Control	What it does
Spoken feedback toggle	Turns spoken feedback on or off.
Provider	Selects the TTS provider. Defaults to the bundled Linux system TTS.
Voice	Selects the voice for the chosen provider.

📸 Screenshot needed: The spoken-feedback section showing the toggle, the provider selector (defaulting to the bundled Linux system TTS), and the per-provider voice selector.

Related pages

• History — searching and managing stored transcriptions.
• Memory — how long-term memory extraction and recall work.

---

Changelog

Date	Change
2026-06-17	Initial version.