Settings

{#if openSection === 'audio'}

Microphone

{#if audioDevicesError}

{audioDevicesError}

{:else if visibleAudioDevices.length === 0}

No input devices detected. Check that a microphone is connected and PulseAudio/PipeWire is running.

{:else}

Auto mode tries the system default first, then any other real input. Speaker-monitor sources (loopback) are skipped because they record system audio rather than the microphone. If dictation is silent, pick the device explicitly here.

{/if}
{/if}
{#if openSection === 'vocabulary'}

Words and phrases the AI cleanup pass should preserve exactly. Useful for medication names, place names, jargon, names of people in your support network, anything Whisper tends to mishear. Vocabulary is scoped to a profile — switch profiles to swap whole vocabularies.

Active Profile

{#if showNewVocabProfile}
{ if (e.key === 'Enter') createVocabProfile(); }} class="bg-bg border border-border rounded-lg px-3 py-2 text-[13px] text-text focus:outline-none focus:border-accent" />
{/if}
{#if profilesStore.active}

Profile Name

{ if (e.key === 'Enter') e.target.blur(); }} class="bg-bg-input border border-border rounded-lg px-3 py-2 text-[13px] text-text min-w-[280px] focus:outline-none focus:border-accent focus:shadow-[0_0_0_3px_rgba(232,168,124,0.1)]" />

Initial Prompt

Passed to Whisper before every transcription in this profile. Used as a biasing prompt — keep it short and topic-specific (names, domain terms, tone cues).

{#if initialPromptDirty} Unsaved changes — saves on blur. {:else} Saved. {/if}

Terms

{ if (e.key === 'Enter') addVocabTerm(); }} class="flex-1 bg-bg-input border border-border rounded-lg px-3 py-2 text-[13px] text-text focus:outline-none focus:border-accent focus:shadow-[0_0_0_3px_rgba(232,168,124,0.1)]" /> { if (e.key === 'Enter') addVocabTerm(); }} class="flex-1 bg-bg-input border border-border rounded-lg px-3 py-2 text-[13px] text-text focus:outline-none focus:border-accent focus:shadow-[0_0_0_3px_rgba(232,168,124,0.1)]" />
{#if !showBulkVocab} {:else}
{/if}
{#if vocabularyError}

{vocabularyError}

{/if} {#if vocabulary.length === 0}

No terms yet. Add one above.

{:else}
    {#each vocabulary as entry (entry.id)}
  • {entry.term}
    {#if entry.note}
    {entry.note}
    {/if}
    • {/each}
{/if} {:else}

Loading profiles…

{/if}
{/if}
{#if openSection === 'transcription'}

Engine

{settings.engine === "whisper" ? "Whisper with the currently shipped English-only models in this build" : "Parakeet CTC 0.6B — English-only, fast when the model is installed"}

Format Mode

{settings.formatMode === "Raw" ? "Exact Whisper output, no formatting" : settings.formatMode === "Clean" ? "Grouped into paragraphs, punctuation tidied" : "Structured with lists, headings, and sections"}

{#if settings.engine === "whisper"}

Whisper Model

{modelDescriptions[settings.modelSize]}

{#if isModelLoaded(settings.modelSize)} Model loaded {:else if isModelDownloaded(settings.modelSize)} Downloaded {:else if downloadingModel === whisperModelId(settings.modelSize)} {downloadProgress}% downloading... {:else} {/if}
{:else}

Parakeet Model

Parakeet CTC 0.6B (int8) — ~613MB, near-instant transcription

{#if parakeetOk} Model loaded {:else if parakeetDownloaded} Downloaded {:else if parakeetDownloading} {parakeetProgress}% downloading... {:else} {/if}
{/if}

Compute Device

{#if hasGpuAcceleration()}

Only accelerators built into this binary are shown here.

{:else}
This build is CPU-only. GPU controls appear in GPU-enabled builds.
{/if}

Language

{#if currentModelIsEnglishOnly()} {:else} {/if} {#if settings.language === "en"} {/if}

{#if currentModelIsEnglishOnly()} The selected model only supports English in this build. {:else} Auto-detect is only meaningful with multilingual models. {/if}

{/if}
{#if openSection === 'processing'}
{/if}
{#if openSection === 'rituals'}

All off by default. Rituals only appear when you ask for them.

{#if settings.ritualsMorning}

ADHD sleep inertia is intense for the first 30–45 minutes after waking. Pick a time when you're genuinely ready to decide.

{/if} {#if settings.ritualsEvening}
{/if}

Launch Corbie at login

So Corbie is already there at the start of the day. Uses your OS's standard autostart — no background tricks.

{#if autostartSyncing}

Updating…

{/if}
{/if}
{#if openSection === 'tasks'}

How the Tasks page surfaces progress. Always additive.

{#if settings.showMomentumSparkline}

Sparkline range

How far back the sparkline looks. Wider windows show longer-arc trends; the narrower window stays close to right-now.

{/if}
{/if}
{#if openSection === 'nudges'}

Gentle, anticipatory reminders. Capped at 3 per hour. Never fires while you're looking at Corbie.

{#if settings.nudgesEnabled}
{/if}
{/if}
{#if openSection === 'rules'}
{/if}
{#if openSection === 'readAloud'}

Uses your operating system's built-in voices. No audio leaves the machine.

{#if ttsVoicesError}

{ttsVoicesError}

{:else if ttsVoicesLoaded && ttsVoices.length === 0}

No additional voices reported by the system synth. Install extra voices through your OS accessibility settings.

{/if}
Slower Normal Faster
{/if}
{#if openSection === 'ai'}

Local LLM for transcript cleanup, smart task extraction, and task breakdown. Runs fully offline after the model is downloaded.

Feature Tier

{settings.aiTier === "off" ? "No local LLM calls. Kon falls back to the existing rule-based path." : settings.aiTier === "cleanup" ? "Use the local model for transcript cleanup and formatting." : "Use the local model for cleanup, task extraction, and task breakdown."}

Model Tier

{#each LLM_MODELS as model} {/each}

{LLM_MODELS.find((model) => model.id === selectedLlmModelId())?.subtitle || "Local model"}

{llmStatus}

{#if llmTestHint}

{llmTestHint}

{/if}
{#if llmDownloadingModel === selectedLlmModelId()} {llmDownloadProgress}% downloading… {:else if !llmModelDownloaded(selectedLlmModelId())} {:else if !llmModelLoaded(selectedLlmModelId())} {:else} {/if}

Recommended for this machine: {LLM_MODELS.find((model) => model.id === (settings.llmModelId || "qwen3_4b_instruct_2507"))?.subtitle || "Qwen3 4B Instruct 2507"} {#if systemInfo} · {Math.round((systemInfo.ram_mb || 0) / 1024)} GB RAM detected {/if}

Cleanup preset

{#if settings.llmPromptPreset === "email"} Formats output as an email paragraph — tight sentences, no markdown, no auto-added greeting or signoff. {:else if settings.llmPromptPreset === "notes"} Formats action items as a bullet list led by imperative verbs; keeps prose informational sentences as prose. {:else if settings.llmPromptPreset === "code"} Preserves technical terms, variable names, file paths, and symbols exactly as spoken. No translation of identifiers. {:else} No preset — the active profile's prompt governs tone alone. {/if}

GPU concurrency

{#if settings.aiGpuConcurrency === "sequential"} On tight-VRAM cards (≤6 GB), loading Whisper + LLM together OOMs. Sequential mode frees the other model before loading; adds a small reload pause between transcribe and cleanup. {:else} Both models stay resident in GPU memory. Faster transitions, but needs enough VRAM to hold both at once. {/if}

{/if}
{#if openSection === 'profiles'}

Custom vocabulary to improve transcription accuracy

{#if showProfiles}
{#each profiles as profile, i}
{profile.name} {profileWordCount(profile.words)} words
{#if editingProfile === i}
{/if}
{/each} {#if showNewProfile}
e.key === "Enter" && createProfile()} data-no-transition />
{:else} {/if}
{/if}

Structured formats for dictation sessions

{#if showTemplates}
{#each templates as template, i}
{template.name} {template.sections.length} sections
{#if editingTemplate === i}
{/if}
{/each} {#if showNewTemplate}
e.key === "Enter" && createTemplate()} data-no-transition />
{:else} {/if}
{/if}
{/if}
{#if openSection === 'output'}
{#if settings.meetingAutoCapture}

Apps to watch (comma-separated)

{ const raw = event.currentTarget.value; settings.meetingAutoCaptureApps = raw .split(",") .map((entry) => entry.trim().toLowerCase()) .filter((entry) => entry.length > 0); }} />
{/if} {#if settings.soundCues}
{Math.round(settings.soundCueVolume * 100)}%
{/if} {#if settings.saveAudio}

Output Folder

{outputFolderPreview}

{#if settings.outputFolder} {/if}
{/if}
{/if}
{#if openSection === 'hotkey'}

Toggle recording from anywhere. Click to change.

{/if}
{#if openSection === 'appearance'}

Theme

Zone

Font Size {settings.fontSize}px

{$_("settings.language")}

{#each SUPPORTED_LOCALES as option} {/each}

{$_("settings.languageDescription")}

{/if}
{#if openSection === 'accessibility'}
{/if}
{#if openSection === 'about'}
{engineStatus}
{#each [ "100% offline — all processing on your machine", "No Python required — compiled Whisper engine", "No cloud — audio never leaves your computer", "No accounts — no sign-up, no tracking", "No telemetry — zero data collection", ] as item}

{item}

{/each}

Kon v1.0 · Powered by whisper.cpp · Built by CORBEL Ltd

Diagnostics

If something goes wrong, generate a diagnostic report to share with the developer. The report contains your settings, recent errors, any crash dumps, and the tail of the log file. Nothing is sent automatically — you preview it first, then choose copy or save.

{#if diagnosticReport} {/if}
{#if diagnosticReportError}

{diagnosticReportError}

{/if} {#if diagnosticReportSavedTo}

{diagnosticReportSavedTo}

{/if} {#if diagnosticReport}
Preview report ({diagnosticReport.length} chars) 
{diagnosticReport}
{/if}

Onboarding

Replay the first-run welcome and the morning / evening / autostart prompts. Your downloaded models, transcripts, and tasks are kept.

{/if}