Shortcut to toggle filler-word removal on/off during or right after recording

[Saisreenivas]

Problem

Filler-word removal is currently an all-or-nothing setting, but the right behavior actually depends on what the dictation is for, and that can change dictation-to-dictation:

• DMing/messaging someone → filler words ("um", "uh", "like") should be stripped, sounds unprofessional otherwise.
• Other use cases (journaling, brainstorming, transcribing a conversation verbatim, etc.) → filler words are wanted/needed, e.g. to preserve exact phrasing or tone.

Right now the only way to change this is to dig into settings between recordings, which doesn't fit how often the use case actually switches — often mid-session, sometimes the user wants to decide after seeing the transcript rather than committing upfront.

Proposed solution

A keyboard shortcut to toggle filler-word removal on/off:

• Usable while a recording is active, not just from a settings screen.
• Ideally also usable right after a recording finishes (post-recording confirmation), for the case where the user wants to decide once they see how the transcript reads, before it's finalized/pasted.
• Current on/off state should be visible somewhere obvious (menu bar icon state, HUD, etc.) so it's clear which mode is active before dictating.

This turns filler-word handling into a per-recording decision instead of a static global setting, which better matches how usage actually varies (e.g. Slack DM vs. a long-form note) throughout a single session.

Alternatives considered

• A static global on/off setting (today's behavior) — doesn't handle the case where the need changes within the same session/day.
• A per-app default (e.g. always strip in Slack, never in Notes) — could help but doesn't cover cases where the user wants to override the default for one specific dictation regardless of app.