perf(asr): speed up local Flow dictation and land CLM/keyboard refactor
Reduce perceived latency from key release to final text: - Adaptive chunking: 2.5s first chunk + 5s follow-ups so short utterances start on-device recognition while still recording. - Session-level ASR warmup and audio-format cache reuse to remove per-utterance cold-start of SpeechAnalyzer. - Mirror live pipelined partials to the keyboard transcript line via a new flow.transcriptionPartial App Group key + Darwin ping. Also commits the accumulated custom language model, Flow session, keyboard extension restructure, and Xiaomi MiMo provider work in progress on this branch.
This commit is contained in:
Rocky
@@ -141,9 +141,6 @@ public struct KeyboardRootView: View {
|
||||
flowSessionActive: state.flowSessionActive,
|
||||
micDisabled: state.micDisabled,
|
||||
micDisabledHint: state.micDisabledHint,
|
||||
isLocalEngine: state.isLocalEngine,
|
||||
localModelsReady: state.localModelsReady,
|
||||
localModelsLoaded: state.localModelsLoaded,
|
||||
cursorDragHintActive: state.cursorDragActive,
|
||||
openSettings: state.openSettings,
|
||||
startFlowSession: state.startFlowSession
|
||||
@@ -337,9 +334,6 @@ private struct TranscriptLine: View {
|
||||
let flowSessionActive: Bool
|
||||
let micDisabled: Bool
|
||||
let micDisabledHint: String
|
||||
let isLocalEngine: Bool
|
||||
let localModelsReady: Bool
|
||||
let localModelsLoaded: Bool
|
||||
let cursorDragHintActive: Bool
|
||||
let openSettings: () -> Void
|
||||
let startFlowSession: () -> Void
|
||||
@@ -367,22 +361,6 @@ private struct TranscriptLine: View {
|
||||
.foregroundStyle(palette.warning)
|
||||
.lineLimit(1)
|
||||
.truncationMode(.tail)
|
||||
} else if isLocalEngine, !localModelsReady {
|
||||
Button(action: openSettings) {
|
||||
HStack(spacing: 4) {
|
||||
Text(ExtL10n.string("keyboard.models.notDownloaded"))
|
||||
Image(systemName: "chevron.right")
|
||||
.font(.system(size: 10, weight: .semibold))
|
||||
}
|
||||
.font(TypeStyle.caption)
|
||||
.foregroundStyle(palette.warning)
|
||||
.lineLimit(1)
|
||||
.truncationMode(.tail)
|
||||
.frame(maxWidth: .infinity)
|
||||
.contentShape(Rectangle())
|
||||
}
|
||||
.buttonStyle(.plain)
|
||||
.accessibilityHint(ExtL10n.text("keyboard.models.downloadHint"))
|
||||
} else if flowSessionActive {
|
||||
ExtL10n.text("keyboard.placeholder.idle")
|
||||
.font(TypeStyle.caption)
|
||||
|
||||
@@ -1,187 +0,0 @@
|
||||
// RecordButton.swift
|
||||
// OSGKeyboard · Keyboard Extension
|
||||
//
|
||||
// Tap-to-toggle mic: tap once to start, tap again to stop. Shows a
|
||||
// remaining-time countdown while recording; last 10 seconds turn red.
|
||||
|
||||
import SwiftUI
|
||||
import OSGKeyboardShared
|
||||
|
||||
struct RecordButton: View {
|
||||
@Environment(\.themePalette) private var palette: ThemePalette
|
||||
|
||||
enum Phase: Equatable {
|
||||
case idle
|
||||
case recording
|
||||
case processing
|
||||
case error
|
||||
}
|
||||
|
||||
let phase: Phase
|
||||
let level: Double // 0...1
|
||||
/// Seconds left in the current utterance; shown only while recording.
|
||||
let remainingSeconds: Int?
|
||||
let isEnabled: Bool
|
||||
let onToggle: () -> Void
|
||||
|
||||
@State private var breath: Bool = false
|
||||
|
||||
init(
|
||||
phase: Phase,
|
||||
level: Double,
|
||||
remainingSeconds: Int? = nil,
|
||||
isEnabled: Bool = true,
|
||||
onToggle: @escaping () -> Void
|
||||
) {
|
||||
self.phase = phase
|
||||
self.level = level
|
||||
self.remainingSeconds = remainingSeconds
|
||||
self.isEnabled = isEnabled
|
||||
self.onToggle = onToggle
|
||||
}
|
||||
|
||||
private var isUrgent: Bool {
|
||||
guard phase == .recording, let remainingSeconds else { return false }
|
||||
return remainingSeconds <= 10
|
||||
}
|
||||
|
||||
/// Decorative rings are sized to stay inside the 121 pt frame applied
|
||||
/// by `KeyboardRootView` so glow / breath animations are not clipped.
|
||||
private enum Layout {
|
||||
static let disc: CGFloat = 95
|
||||
static let outerRing: CGFloat = 106
|
||||
static let breathRing: CGFloat = 100
|
||||
static let glow: CGFloat = 119
|
||||
}
|
||||
|
||||
var body: some View {
|
||||
ZStack {
|
||||
Circle()
|
||||
.stroke(palette.recordRed.opacity(isUrgent ? 0.55 : 0.35), lineWidth: isUrgent ? 3 : 2)
|
||||
.frame(width: Layout.breathRing, height: Layout.breathRing)
|
||||
.scaleEffect(breath ? 1.18 : 0.95)
|
||||
.opacity(phase == .recording ? 1 : 0)
|
||||
.animation(Motion.breath, value: breath)
|
||||
|
||||
Circle()
|
||||
.fill(
|
||||
RadialGradient(
|
||||
colors: [palette.recordRed.opacity(0.55), .clear],
|
||||
center: .center,
|
||||
startRadius: 46,
|
||||
endRadius: 92
|
||||
)
|
||||
)
|
||||
.frame(width: Layout.glow, height: Layout.glow)
|
||||
.opacity(phase == .recording ? 0.4 + level * 0.6 : 0)
|
||||
.blur(radius: 18)
|
||||
.animation(Motion.soft, value: phase)
|
||||
.animation(Motion.soft, value: level)
|
||||
|
||||
Circle()
|
||||
.stroke(
|
||||
Color.white.opacity(phase == .idle ? 0.08 : 0.12),
|
||||
lineWidth: 0.5
|
||||
)
|
||||
.frame(width: Layout.outerRing, height: Layout.outerRing)
|
||||
|
||||
ZStack {
|
||||
Circle()
|
||||
.fill(discGradient)
|
||||
Circle()
|
||||
.stroke(Color.white.opacity(0.16), lineWidth: 1)
|
||||
.blendMode(.overlay)
|
||||
|
||||
Group {
|
||||
switch phase {
|
||||
case .idle:
|
||||
Image(systemName: "mic.fill")
|
||||
.font(.system(size: 36, weight: .medium))
|
||||
.foregroundStyle(.white)
|
||||
case .recording:
|
||||
VStack(spacing: 3) {
|
||||
if let remainingSeconds {
|
||||
Text(formatRemaining(remainingSeconds))
|
||||
.font(.system(size: 22, weight: .semibold, design: .rounded))
|
||||
.foregroundStyle(.white)
|
||||
.monospacedDigit()
|
||||
.contentTransition(.numericText())
|
||||
// 倒计时略下移，与波形一起在圆盘内更居中。
|
||||
.offset(y: 3)
|
||||
}
|
||||
WaveformView(
|
||||
level: level,
|
||||
color: Color(red: 1.0, green: 0.78, blue: 0.78),
|
||||
active: true
|
||||
)
|
||||
.frame(width: 73, height: 32)
|
||||
.opacity(0.4)
|
||||
.scaleEffect(0.96)
|
||||
}
|
||||
.transition(.opacity)
|
||||
case .processing:
|
||||
ProgressView()
|
||||
.progressViewStyle(.circular)
|
||||
.tint(palette.textPrimary)
|
||||
.scaleEffect(1.25)
|
||||
case .error:
|
||||
Image(systemName: "exclamationmark.triangle.fill")
|
||||
.font(.system(size: 32, weight: .medium))
|
||||
.foregroundStyle(palette.warning)
|
||||
}
|
||||
}
|
||||
}
|
||||
.frame(width: Layout.disc, height: Layout.disc)
|
||||
.animation(Motion.soft, value: phase)
|
||||
.animation(Motion.soft, value: remainingSeconds)
|
||||
}
|
||||
.contentShape(Circle())
|
||||
.opacity(isEnabled ? 1 : 0.45)
|
||||
.onTapGesture {
|
||||
guard isEnabled, phase != .processing else { return }
|
||||
onToggle()
|
||||
}
|
||||
.onAppear { breath = (phase == .recording) }
|
||||
.onChange(of: phase) { _, new in
|
||||
breath = (new == .recording)
|
||||
}
|
||||
.accessibilityLabel(ExtL10n.text("keyboard.tapToTalkA11y"))
|
||||
}
|
||||
|
||||
private func formatRemaining(_ seconds: Int) -> String {
|
||||
let m = seconds / 60
|
||||
let s = seconds % 60
|
||||
return String(format: "%d:%02d", m, s)
|
||||
}
|
||||
|
||||
private var discGradient: LinearGradient {
|
||||
switch phase {
|
||||
case .recording:
|
||||
let colors: [Color] = isUrgent
|
||||
? [palette.recordRed, palette.recordRed.opacity(0.85)]
|
||||
: [palette.recordRed.opacity(0.95), palette.recordRed.opacity(0.75)]
|
||||
return LinearGradient(colors: colors, startPoint: .top, endPoint: .bottom)
|
||||
case .processing:
|
||||
return LinearGradient(
|
||||
colors: [palette.surfaceElevated, palette.surface],
|
||||
startPoint: .top,
|
||||
endPoint: .bottom
|
||||
)
|
||||
case .error:
|
||||
return LinearGradient(
|
||||
colors: [palette.warning.opacity(0.85), palette.warning.opacity(0.55)],
|
||||
startPoint: .top,
|
||||
endPoint: .bottom
|
||||
)
|
||||
case .idle:
|
||||
return LinearGradient(
|
||||
colors: [
|
||||
palette.accent.opacity(0.95),
|
||||
palette.accent.opacity(0.75)
|
||||
],
|
||||
startPoint: .top,
|
||||
endPoint: .bottom
|
||||
)
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -1,148 +0,0 @@
|
||||
// TranslationChip.swift
|
||||
// OSGKeyboard · Keyboard Extension
|
||||
//
|
||||
// Compact chip rendered to the right of `LocaleChip` on the keyboard
|
||||
// top bar. Doubles as both the on/off switch and the target-language
|
||||
// picker — same Menu pattern as `LocaleChip` so muscle memory transfers.
|
||||
//
|
||||
// v0.2.1 follow-up: removed the explicit on/off toggle entry. The
|
||||
// chip is now a pure picker over the 11 catalog rows (off + 10
|
||||
// locales); selecting "不翻译" turns translation off, selecting any
|
||||
// locale turns it on with that target. `translationEnabled` is
|
||||
// derived from the locale id so the chip / pipeline read the same
|
||||
// source of truth.
|
||||
//
|
||||
// v0.2.1 final review: dropped the "needs cloud" warning state —
|
||||
// both engines now run the translate-and-polish step (the local
|
||||
// engine routes through DeepSeek via
|
||||
// `ProviderConfig.localModeProviderId`). The chip is therefore just
|
||||
// off / on, with the same accent treatment either way.
|
||||
//
|
||||
// Visual states:
|
||||
// • off → dim outline, "翻译" chip label (menu first row = "不翻译")
|
||||
// • on (any engine) → accent fill, "→ EN" / "→ 日本語" style label
|
||||
//
|
||||
// Stays in the same visual family as `CloudEngineChip` / `LocaleChip`
|
||||
// (Capsule + 28 pt min height + 6 pt vertical padding) so the top bar
|
||||
// doesn't grow when translation is enabled.
|
||||
|
||||
import SwiftUI
|
||||
import OSGKeyboardShared
|
||||
|
||||
struct TranslationChip: View, Equatable {
|
||||
/// Passed in as a value (not read from `@Environment`) so the chip can
|
||||
/// be wrapped in `.equatable()` at the call site: `EquatableView`
|
||||
/// suppresses environment-driven refreshes, so injecting the palette
|
||||
/// here keeps colours correct across dark/light switches.
|
||||
let palette: ThemePalette
|
||||
/// The active target-locale id (`offLocaleId` == translation off).
|
||||
let targetLocaleId: String
|
||||
/// Writes the picked locale id — wired to `state.setTranslationTargetLocaleId`.
|
||||
let onSelect: (String) -> Void
|
||||
|
||||
/// Only `palette` and `targetLocaleId` drive the visuals; the
|
||||
/// `onSelect` closure is deliberately excluded from equality. Because
|
||||
/// the keyboard polls the App Group at 1 Hz (each poll re-publishes the
|
||||
/// `KeyboardState`), the parent view re-renders every second. Without
|
||||
/// this, SwiftUI would rebuild the `Menu` on every poll — dismissing an
|
||||
/// open picker or snapping its scroll position back to the top. With
|
||||
/// `.equatable()` the picker is rebuilt only on a real state change.
|
||||
nonisolated static func == (lhs: TranslationChip, rhs: TranslationChip) -> Bool {
|
||||
lhs.palette == rhs.palette && lhs.targetLocaleId == rhs.targetLocaleId
|
||||
}
|
||||
|
||||
var body: some View {
|
||||
Menu {
|
||||
// v0.2.1 follow-up: pure picker over the full catalog,
|
||||
// including `offLocaleId` at the top so "turn off" is one
|
||||
// tap from any enabled state. Picking a row writes
|
||||
// `translationTargetLocaleId`; `translationEnabled` is
|
||||
// derived from it.
|
||||
ForEach(TranslationLanguageCatalog.all) { language in
|
||||
Button {
|
||||
onSelect(language.id)
|
||||
} label: {
|
||||
if language.id == targetLocaleId {
|
||||
Label(displayLabel(for: language), systemImage: "checkmark")
|
||||
} else {
|
||||
Text(displayLabel(for: language))
|
||||
}
|
||||
}
|
||||
}
|
||||
} label: {
|
||||
label
|
||||
}
|
||||
.menuStyle(.button)
|
||||
.accessibilityLabel(ExtL10n.text("keyboard.translation.a11y"))
|
||||
.accessibilityHint(ExtL10n.text("keyboard.translation.a11yHint"))
|
||||
}
|
||||
|
||||
@ViewBuilder
|
||||
private var label: some View {
|
||||
let target = TranslationLanguageCatalog.resolve(targetLocaleId)
|
||||
let enabled = targetLocaleId != TranslationLanguageCatalog.offLocaleId
|
||||
|
||||
HStack(spacing: 4) {
|
||||
Image(systemName: enabled ? "character.bubble" : "character.bubble.fill")
|
||||
Text(chipLabel(target: target, enabled: enabled))
|
||||
Image(systemName: "chevron.down")
|
||||
.font(.system(size: 8, weight: .bold))
|
||||
}
|
||||
.font(TypeStyle.caption2)
|
||||
.foregroundStyle(foreground(enabled: enabled))
|
||||
.padding(.horizontal, Spacing.xs + 2)
|
||||
.padding(.vertical, 6)
|
||||
.frame(minHeight: 28)
|
||||
.background(background(enabled: enabled), in: Capsule())
|
||||
.overlay(Capsule().stroke(stroke(enabled: enabled), lineWidth: 0.5))
|
||||
}
|
||||
|
||||
private func displayLabel(for language: TranslationLanguage) -> String {
|
||||
if language.id == TranslationLanguageCatalog.offLocaleId {
|
||||
return ExtL10n.string("keyboard.translation.offMenu")
|
||||
}
|
||||
return language.nativeName
|
||||
}
|
||||
|
||||
private func chipLabel(target: TranslationLanguage, enabled: Bool) -> String {
|
||||
if !enabled {
|
||||
return ExtL10n.string("keyboard.translation.chip")
|
||||
}
|
||||
// Short form: "→EN" / "→日" style. Falls back to the prompt
|
||||
// language name for languages without a chip-style abbreviation
|
||||
// (e.g. French → "FR" via the 2-letter prefix).
|
||||
let short = shortLabel(for: target)
|
||||
return "→\(short)"
|
||||
}
|
||||
|
||||
private func shortLabel(for target: TranslationLanguage) -> String {
|
||||
switch target.id {
|
||||
case "en": return "EN"
|
||||
case "zh-Hans": return "中"
|
||||
case "zh-Hant": return "繁"
|
||||
case "ja": return "日"
|
||||
case "ko": return "韩"
|
||||
case "fr": return "FR"
|
||||
case "de": return "DE"
|
||||
case "es": return "ES"
|
||||
case "ru": return "RU"
|
||||
case "pt": return "PT"
|
||||
default: return target.promptLanguageName
|
||||
}
|
||||
}
|
||||
|
||||
private func foreground(enabled: Bool) -> Color {
|
||||
if enabled { return palette.accent }
|
||||
return palette.textPrimary
|
||||
}
|
||||
|
||||
private func background(enabled: Bool) -> Color {
|
||||
if enabled { return palette.accent.opacity(0.15) }
|
||||
return palette.surfaceElevated
|
||||
}
|
||||
|
||||
private func stroke(enabled: Bool) -> Color {
|
||||
if enabled { return palette.accent.opacity(0.35) }
|
||||
return palette.divider
|
||||
}
|
||||
}
|
||||
@@ -1,60 +0,0 @@
|
||||
// WaveformView.swift
|
||||
// OSGKeyboard · Keyboard Extension
|
||||
//
|
||||
// Symmetric, real-time driven waveform. 18 bars centred around a vertical
|
||||
// axis. The dominant bar is driven by the current RMS; surrounding bars
|
||||
// decay on a small position-based curve so the visual feels like a
|
||||
// horizontal speaker cone, not random noise.
|
||||
|
||||
import SwiftUI
|
||||
import OSGKeyboardShared
|
||||
|
||||
struct WaveformView: View {
|
||||
@Environment(\.themePalette) private var palette: ThemePalette
|
||||
|
||||
let level: Double // 0...1, smoothed RMS
|
||||
let barCount: Int
|
||||
let color: Color?
|
||||
let active: Bool // when false, bars collapse to a thin resting line
|
||||
|
||||
init(
|
||||
level: Double,
|
||||
barCount: Int = 18,
|
||||
color: Color? = nil,
|
||||
active: Bool = true
|
||||
) {
|
||||
self.level = max(0, min(1, level))
|
||||
self.barCount = barCount
|
||||
self.color = color
|
||||
self.active = active
|
||||
}
|
||||
|
||||
private var resolvedColor: Color {
|
||||
color ?? palette.recordRed
|
||||
}
|
||||
|
||||
var body: some View {
|
||||
TimelineView(.animation(minimumInterval: 1.0 / 30.0)) { context in
|
||||
HStack(alignment: .center, spacing: 3) {
|
||||
ForEach(0..<barCount, id: \.self) { i in
|
||||
Capsule()
|
||||
.fill(resolvedColor)
|
||||
.frame(width: 2.4, height: height(for: i, time: context.date.timeIntervalSinceReferenceDate))
|
||||
.opacity(active ? 1.0 : 0.45)
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
private func height(for index: Int, time: TimeInterval) -> CGFloat {
|
||||
guard active else { return 4 }
|
||||
let centre = Double(barCount - 1) / 2.0
|
||||
let distance = abs(Double(index) - centre) / max(centre, 1)
|
||||
// Per-bar small wobble so the line is alive but tied to level.
|
||||
let phase = sin(time * 4.0 + Double(index) * 0.45)
|
||||
let wobble = 0.18 * phase
|
||||
let magnitude = max(0, min(1, Double(level) + wobble))
|
||||
let profile = 1.0 - pow(distance, 1.4) * 0.85
|
||||
return CGFloat(max(6, 32 * magnitude * profile))
|
||||
}
|
||||
}
|
||||
Reference in New Issue
Block a user