feat(capture): dictation, personalities, 0.5.0

Ships the Capture release end to end. Global-hotkey dictation with
synthetic paste into the focused app on macOS and Windows, an on-screen
pill across recording / transcribing / refining, customizable push-to-
talk and toggle chords, and an accessibility-permission prompt scoped to
Settings → Captures with inline re-check feedback.

Voice profiles gain optional personalities that power compose / rewrite /
respond actions via a local Qwen3 LLM — shared with refinement, so there
is one local LLM in the app, not two.

Refinement hardened with deterministic Whisper-loop collapse before the
LLM sees the transcript, per-capture flag snapshots for re-runs, and a
ten-transcript evaluation harness across every bundled refinement size.

Version bump 0.4.5 → 0.5.0.

Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>
This commit is contained in:
James Pine
2026-04-22 18:49:16 -07:00
co-authored by Claude Opus 4.7
parent ed2eec591a
commit 87c582ad54
84 changed files with 11043 additions and 512 deletions
@@ -1,4 +1,5 @@
import { Loader2, Mic } from 'lucide-react';
import { useMutation } from '@tanstack/react-query';
import { Loader2, Mic, RefreshCw, Sparkles } from 'lucide-react';
import { useEffect } from 'react';
import { Button } from '@/components/ui/button';
import { Card, CardContent, CardHeader, CardTitle } from '@/components/ui/card';
@@ -20,6 +21,8 @@ import {
SelectValue,
} from '@/components/ui/select';
import { Textarea } from '@/components/ui/textarea';
import { useToast } from '@/components/ui/use-toast';
import { apiClient } from '@/lib/api/client';
import { getLanguageOptionsForEngine, type LanguageCode } from '@/lib/constants/languages';
import { useGenerationForm } from '@/lib/hooks/useGenerationForm';
import { useProfile } from '@/lib/hooks/useProfiles';
@@ -41,6 +44,7 @@ function getEngineSelectValue(engine: string): string {
export function GenerationForm() {
const selectedProfileId = useUIStore((state) => state.selectedProfileId);
const { data: selectedProfile } = useProfile(selectedProfileId || '');
const { toast } = useToast();
const { form, handleSubmit, isPending } = useGenerationForm();
@@ -63,6 +67,51 @@ export function GenerationForm() {
await handleSubmit(data, selectedProfileId);
}
// ── Personality-driven text generation ─────────────────────────────
// Compose fills the empty textarea with a fresh in-character line.
// Rewrite restates whatever's in the textarea in the profile's voice.
// Both buttons hide entirely when the selected profile has no
// personality set — nothing to drive the LLM with otherwise.
const personality = selectedProfile?.personality?.trim() || '';
const hasPersonality = personality.length > 0;
const currentText = form.watch('text');
const textHasContent = (currentText || '').trim().length > 0;
const composeMutation = useMutation({
mutationFn: async () => {
if (!selectedProfileId) throw new Error('No profile selected');
return apiClient.composeWithPersonality(selectedProfileId);
},
onSuccess: (result) => {
form.setValue('text', result.text, { shouldDirty: true, shouldValidate: true });
},
onError: (err: Error) => {
toast({
title: 'Compose failed',
description: err.message || 'Could not generate text from this personality.',
variant: 'destructive',
});
},
});
const rewriteMutation = useMutation({
mutationFn: async (text: string) => {
if (!selectedProfileId) throw new Error('No profile selected');
return apiClient.rewriteWithPersonality(selectedProfileId, text);
},
onSuccess: (result) => {
form.setValue('text', result.text, { shouldDirty: true, shouldValidate: true });
},
onError: (err: Error) => {
toast({
title: 'Rewrite failed',
description: err.message || 'Could not rewrite the text in this voice.',
variant: 'destructive',
});
},
});
return (
<Card>
<CardHeader>
@@ -118,6 +167,52 @@ export function GenerationForm() {
)}
/>
{hasPersonality && (
<div className="flex items-center gap-2">
<Button
type="button"
variant="outline"
size="sm"
disabled={
!selectedProfileId ||
textHasContent ||
composeMutation.isPending ||
rewriteMutation.isPending
}
onClick={() => composeMutation.mutate()}
>
{composeMutation.isPending ? (
<Loader2 className="mr-2 h-3.5 w-3.5 animate-spin" />
) : (
<Sparkles className="mr-2 h-3.5 w-3.5" />
)}
Compose
</Button>
<Button
type="button"
variant="outline"
size="sm"
disabled={
!selectedProfileId ||
!textHasContent ||
rewriteMutation.isPending ||
composeMutation.isPending
}
onClick={() => rewriteMutation.mutate(currentText || '')}
>
{rewriteMutation.isPending ? (
<Loader2 className="mr-2 h-3.5 w-3.5 animate-spin" />
) : (
<RefreshCw className="mr-2 h-3.5 w-3.5" />
)}
Rewrite in voice
</Button>
<span className="text-xs text-muted-foreground">
Uses the profile's personality.
</span>
</div>
)}
{form.watch('engine') === 'qwen_custom_voice' && (
<FormField
control={form.control}