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
@@ -0,0 +1,262 @@
|
||||
// AppGroupConfiguration.swift
|
||||
// OSGKeyboard · Shared
|
||||
//
|
||||
// Single source of truth for App Group UserDefaults keys (`config.*`).
|
||||
// Both `ProviderConfig` (main app) and `AppGroupStore` (keyboard ext)
|
||||
// should read/write through this type so keys and defaults stay aligned.
|
||||
|
||||
import Foundation
|
||||
|
||||
public struct AppGroupConfiguration: Sendable, Equatable {
|
||||
// MARK: - Keys
|
||||
|
||||
public enum Keys {
|
||||
public static let providerId = "config.providerId"
|
||||
public static let baseURL = "config.baseURL"
|
||||
/// Legacy plaintext slot — migrated to Keychain on first read.
|
||||
public static let apiKeyLegacy = "config.apiKey"
|
||||
public static let model = "config.model"
|
||||
public static let modeId = "config.modeId"
|
||||
public static let localeId = "config.localeId"
|
||||
public static let engineMode = "config.engineMode"
|
||||
public static let hasCompletedOnboarding = "config.hasCompletedOnboarding"
|
||||
public static let onboardingPage = "config.onboardingPage"
|
||||
public static let hasAcknowledgedCloudSharing = "config.hasAcknowledgedCloudSharing"
|
||||
public static let uiLanguage = "config.uiLanguage"
|
||||
public static let translationTargetLocaleId = "config.translationTargetLocaleId"
|
||||
public static let handednessPreference = "config.handednessPreference"
|
||||
public static let cursorDragNavigationEnabled = "config.cursorDragNavigationEnabled"
|
||||
public static let polishIntensity = "config.polishIntensity"
|
||||
public static let detectedAppContext = "config.detectedAppContext"
|
||||
public static let detectedAppContextAt = "config.detectedAppContextAt"
|
||||
public static let personalDictionary = "config.personalDictionary.v1"
|
||||
}
|
||||
|
||||
// MARK: - Stored fields
|
||||
|
||||
public var providerId: String
|
||||
public var baseURL: String
|
||||
public var model: String
|
||||
public var modeId: String
|
||||
public var localeId: String
|
||||
public var engineMode: String
|
||||
public var hasCompletedOnboarding: Bool
|
||||
public var onboardingPage: Int
|
||||
public var hasAcknowledgedCloudSharing: Bool
|
||||
public var uiLanguage: AppUILanguage
|
||||
public var translationTargetLocaleId: String
|
||||
public var handednessPreference: HandednessPreference
|
||||
public var cursorDragNavigationEnabled: Bool
|
||||
public var polishIntensity: PolishIntensity
|
||||
public var personalDictionary: PersonalDictionary
|
||||
|
||||
// MARK: - Derived
|
||||
|
||||
/// Translation is on iff a target locale other than `offLocaleId` is selected.
|
||||
public var translationEnabled: Bool {
|
||||
translationTargetLocaleId != TranslationLanguageCatalog.offLocaleId
|
||||
}
|
||||
|
||||
public var isTranslationEffective: Bool {
|
||||
translationEnabled
|
||||
}
|
||||
|
||||
public var isLocalEngine: Bool {
|
||||
engineMode == "local"
|
||||
}
|
||||
|
||||
public var polishModeForPipeline: PolishingService.PolishMode {
|
||||
isTranslationEffective
|
||||
? .translate(targetLocaleId: translationTargetLocaleId)
|
||||
: .polish
|
||||
}
|
||||
|
||||
/// Local engine pins the LLM step to DeepSeek; cloud uses the user's provider.
|
||||
public var polishProviderIdOverride: String? {
|
||||
engineMode == "local" ? "deepseek" : nil
|
||||
}
|
||||
|
||||
public var isCloudAPIKeyMissingForVoiceInput: Bool {
|
||||
guard engineMode == "cloud" else { return false }
|
||||
return apiKey.trimmingCharacters(in: .whitespacesAndNewlines).isEmpty
|
||||
}
|
||||
|
||||
/// API key lives in the Keychain (cross-process, encrypted at rest).
|
||||
public var apiKey: String {
|
||||
Keychain.apiKey(for: providerId) ?? ""
|
||||
}
|
||||
|
||||
public func makeClient() -> LLMClient {
|
||||
OpenAICompatibleClient(
|
||||
baseURL: baseURL,
|
||||
apiKey: apiKey,
|
||||
model: model
|
||||
)
|
||||
}
|
||||
|
||||
// MARK: - Detected app context
|
||||
|
||||
public func detectedAppContext(from defaults: UserDefaults) -> (context: AppContext, observedAt: Date)? {
|
||||
guard let raw = defaults.string(forKey: Keys.detectedAppContext),
|
||||
let value = AppContext(rawValue: raw)
|
||||
else { return nil }
|
||||
let timestamp = defaults.object(forKey: Keys.detectedAppContextAt) as? Date ?? .distantPast
|
||||
return (value, timestamp)
|
||||
}
|
||||
|
||||
public mutating func setDetectedAppContext(_ context: AppContext, at date: Date = Date(), to defaults: UserDefaults) {
|
||||
defaults.set(context.rawValue, forKey: Keys.detectedAppContext)
|
||||
defaults.set(date, forKey: Keys.detectedAppContextAt)
|
||||
}
|
||||
|
||||
// MARK: - Load / save
|
||||
|
||||
/// Loads configuration from App Group defaults. Returns `nil` when the suite is unavailable.
|
||||
public static func load(from defaults: UserDefaults? = nil) -> AppGroupConfiguration? {
|
||||
guard let store = defaults ?? AppGroup.defaultsIfAvailable else { return nil }
|
||||
return load(fromAvailable: store)
|
||||
}
|
||||
|
||||
/// Loads configuration from a known-available UserDefaults suite.
|
||||
public static func load(fromAvailable defaults: UserDefaults) -> AppGroupConfiguration {
|
||||
var config = AppGroupConfiguration(
|
||||
providerId: defaults.string(forKey: Keys.providerId) ?? "openai",
|
||||
baseURL: "",
|
||||
model: "",
|
||||
modeId: defaults.string(forKey: Keys.modeId) ?? "polish",
|
||||
localeId: defaults.string(forKey: Keys.localeId) ?? "auto",
|
||||
engineMode: defaults.string(forKey: Keys.engineMode) ?? "cloud",
|
||||
hasCompletedOnboarding: defaults.bool(forKey: Keys.hasCompletedOnboarding),
|
||||
onboardingPage: {
|
||||
let saved = defaults.integer(forKey: Keys.onboardingPage)
|
||||
return saved > 0 ? saved : 0
|
||||
}(),
|
||||
hasAcknowledgedCloudSharing: defaults.bool(forKey: Keys.hasAcknowledgedCloudSharing),
|
||||
uiLanguage: AppUILanguage.fromStored(defaults.string(forKey: Keys.uiLanguage)),
|
||||
translationTargetLocaleId: defaults.string(forKey: Keys.translationTargetLocaleId)
|
||||
?? TranslationLanguageCatalog.offLocaleId,
|
||||
handednessPreference: HandednessPreference.fromStored(
|
||||
defaults.string(forKey: Keys.handednessPreference)
|
||||
),
|
||||
cursorDragNavigationEnabled: {
|
||||
if defaults.object(forKey: Keys.cursorDragNavigationEnabled) == nil {
|
||||
return true
|
||||
}
|
||||
return defaults.bool(forKey: Keys.cursorDragNavigationEnabled)
|
||||
}(),
|
||||
polishIntensity: resolvePolishIntensity(from: defaults),
|
||||
personalDictionary: decodePersonalDictionary(from: defaults)
|
||||
)
|
||||
|
||||
let preset = LLMProvider.provider(id: config.providerId)
|
||||
if config.baseURL.isEmpty {
|
||||
config.baseURL = defaults.string(forKey: Keys.baseURL) ?? preset.defaultBaseURL
|
||||
}
|
||||
if config.model.isEmpty {
|
||||
config.model = defaults.string(forKey: Keys.model) ?? preset.defaultModel
|
||||
}
|
||||
|
||||
// One-shot legacy migration: plaintext apiKey in UserDefaults → Keychain.
|
||||
_ = resolveAPIKey(defaults: defaults, providerId: config.providerId)
|
||||
|
||||
// Cloud no longer exposes off/transcribe; migrate legacy values.
|
||||
if config.engineMode == "cloud", config.modeId != "polish" {
|
||||
config.modeId = "polish"
|
||||
defaults.set("polish", forKey: Keys.modeId)
|
||||
}
|
||||
// DeepSeek is local-engine only — never a cloud picker choice.
|
||||
if config.engineMode == "cloud", config.providerId == "deepseek" {
|
||||
let openAI = LLMProvider.provider(id: "openai")
|
||||
config.providerId = openAI.id
|
||||
config.baseURL = openAI.defaultBaseURL
|
||||
config.model = openAI.defaultModel
|
||||
defaults.set(openAI.id, forKey: Keys.providerId)
|
||||
defaults.set(openAI.defaultBaseURL, forKey: Keys.baseURL)
|
||||
defaults.set(openAI.defaultModel, forKey: Keys.model)
|
||||
}
|
||||
|
||||
return config
|
||||
}
|
||||
|
||||
public func save(to defaults: UserDefaults) {
|
||||
defaults.set(providerId, forKey: Keys.providerId)
|
||||
defaults.set(baseURL, forKey: Keys.baseURL)
|
||||
defaults.set(model, forKey: Keys.model)
|
||||
defaults.set(modeId, forKey: Keys.modeId)
|
||||
defaults.set(localeId, forKey: Keys.localeId)
|
||||
defaults.set(engineMode, forKey: Keys.engineMode)
|
||||
defaults.set(hasCompletedOnboarding, forKey: Keys.hasCompletedOnboarding)
|
||||
defaults.set(onboardingPage, forKey: Keys.onboardingPage)
|
||||
defaults.set(hasAcknowledgedCloudSharing, forKey: Keys.hasAcknowledgedCloudSharing)
|
||||
defaults.set(uiLanguage.rawValue, forKey: Keys.uiLanguage)
|
||||
defaults.set(translationTargetLocaleId, forKey: Keys.translationTargetLocaleId)
|
||||
defaults.set(handednessPreference.rawValue, forKey: Keys.handednessPreference)
|
||||
defaults.set(cursorDragNavigationEnabled, forKey: Keys.cursorDragNavigationEnabled)
|
||||
defaults.set(polishIntensity.rawValue, forKey: Keys.polishIntensity)
|
||||
Self.encodePersonalDictionary(personalDictionary, to: defaults)
|
||||
}
|
||||
|
||||
// MARK: - Private helpers
|
||||
|
||||
private static func resolvePolishIntensity(from defaults: UserDefaults) -> PolishIntensity {
|
||||
guard let raw = defaults.string(forKey: Keys.polishIntensity) else {
|
||||
return .default
|
||||
}
|
||||
let resolved = PolishIntensity.resolve(storedRawValue: raw)
|
||||
if raw == PolishIntensity.legacyOffRawValue {
|
||||
defaults.set(resolved.rawValue, forKey: Keys.polishIntensity)
|
||||
}
|
||||
return resolved
|
||||
}
|
||||
|
||||
private static func decodePersonalDictionary(from defaults: UserDefaults) -> PersonalDictionary {
|
||||
guard let data = defaults.data(forKey: Keys.personalDictionary) else {
|
||||
return .empty
|
||||
}
|
||||
do {
|
||||
var dictionary = try JSONDecoder().decode(PersonalDictionary.self, from: data)
|
||||
if dictionary.entries.contains(where: { $0.source == .history }) {
|
||||
for index in dictionary.entries.indices where dictionary.entries[index].source == .history {
|
||||
dictionary.entries[index].source = .manual
|
||||
}
|
||||
dictionary.version += 1
|
||||
if let migrated = try? JSONEncoder().encode(dictionary) {
|
||||
defaults.set(migrated, forKey: Keys.personalDictionary)
|
||||
}
|
||||
}
|
||||
return dictionary
|
||||
} catch {
|
||||
OSGLog.config.warning("personalDictionary decode failed: \(error.localizedDescription, privacy: .public)")
|
||||
return .empty
|
||||
}
|
||||
}
|
||||
|
||||
private static func encodePersonalDictionary(_ dictionary: PersonalDictionary, to defaults: UserDefaults) {
|
||||
do {
|
||||
let data = try JSONEncoder().encode(dictionary)
|
||||
defaults.set(data, forKey: Keys.personalDictionary)
|
||||
} catch {
|
||||
OSGLog.config.warning("personalDictionary encode failed: \(error.localizedDescription, privacy: .public)")
|
||||
}
|
||||
}
|
||||
|
||||
/// Read the API key from the Keychain, falling back to a one-time migration from UserDefaults.
|
||||
static func resolveAPIKey(defaults: UserDefaults?, providerId: String) -> String {
|
||||
if let stored = Keychain.apiKey(for: providerId), !stored.isEmpty {
|
||||
return stored
|
||||
}
|
||||
if let legacyKeychain = Keychain.legacyAPIKey(), !legacyKeychain.isEmpty {
|
||||
try? Keychain.setAPIKey(legacyKeychain, for: providerId)
|
||||
try? Keychain.deleteLegacyAPIKey()
|
||||
return legacyKeychain
|
||||
}
|
||||
if let defaults,
|
||||
let legacy = defaults.string(forKey: Keys.apiKeyLegacy),
|
||||
!legacy.isEmpty {
|
||||
try? Keychain.setAPIKey(legacy, for: providerId)
|
||||
defaults.removeObject(forKey: Keys.apiKeyLegacy)
|
||||
return legacy
|
||||
}
|
||||
return ""
|
||||
}
|
||||
}
|
||||
@@ -10,12 +10,11 @@ public enum EngineServiceLabel {
|
||||
engineMode: String,
|
||||
providerId: String,
|
||||
model: String,
|
||||
localASRBackend: LocalASRBackend = .speechAnalyzer,
|
||||
language: AppUILanguage? = nil
|
||||
) -> String {
|
||||
let lang = language ?? AppGroupStore().uiLanguage
|
||||
if engineMode == "local" {
|
||||
let asrName = asrDisplayName(for: localASRBackend, language: lang)
|
||||
let asrName = SharedL10n.string("engine.asr.appleSpeech", language: lang)
|
||||
return SharedL10n.format("engine.summary.local", language: lang, asrName)
|
||||
}
|
||||
let providerName = ProviderDisplayName.name(for: providerId, language: lang)
|
||||
@@ -30,14 +29,4 @@ public enum EngineServiceLabel {
|
||||
trimmedModel
|
||||
)
|
||||
}
|
||||
|
||||
private static func asrDisplayName(
|
||||
for backend: LocalASRBackend,
|
||||
language: AppUILanguage
|
||||
) -> String {
|
||||
// v0.2.0: only the iOS SpeechAnalyzer path remains. We keep the
|
||||
// switch on `LocalASRBackend` so the next non-iOS backend can
|
||||
// slot in without touching every call site.
|
||||
return SharedL10n.string("engine.asr.appleSpeech", language: language)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@@ -6,8 +6,10 @@
|
||||
import Foundation
|
||||
|
||||
public struct FlowUtteranceChunkConfig: Sendable, Equatable {
|
||||
/// Target maximum duration per ASR chunk.
|
||||
public let maxChunkDurationSeconds: TimeInterval
|
||||
/// Target duration for the first ASR chunk (starts pipelining early).
|
||||
public let firstChunkDurationSeconds: TimeInterval
|
||||
/// Target duration for later chunks once pipelining is underway.
|
||||
public let subsequentChunkDurationSeconds: TimeInterval
|
||||
/// Tail overlap fed into the next chunk for boundary dedup when stitching.
|
||||
public let overlapDurationSeconds: TimeInterval
|
||||
/// After hitting the max window, wait up to this long for a pause before hard-splitting.
|
||||
@@ -17,21 +19,52 @@ public struct FlowUtteranceChunkConfig: Sendable, Equatable {
|
||||
public let sampleRate: Int
|
||||
|
||||
public init(
|
||||
maxChunkDurationSeconds: TimeInterval,
|
||||
firstChunkDurationSeconds: TimeInterval = 2.5,
|
||||
subsequentChunkDurationSeconds: TimeInterval = 5.0,
|
||||
overlapDurationSeconds: TimeInterval,
|
||||
pauseExtensionMaxSeconds: TimeInterval,
|
||||
pauseRMSThreshold: Float,
|
||||
sampleRate: Int
|
||||
) {
|
||||
self.maxChunkDurationSeconds = maxChunkDurationSeconds
|
||||
self.firstChunkDurationSeconds = firstChunkDurationSeconds
|
||||
self.subsequentChunkDurationSeconds = subsequentChunkDurationSeconds
|
||||
self.overlapDurationSeconds = overlapDurationSeconds
|
||||
self.pauseExtensionMaxSeconds = pauseExtensionMaxSeconds
|
||||
self.pauseRMSThreshold = pauseRMSThreshold
|
||||
self.sampleRate = sampleRate
|
||||
}
|
||||
|
||||
/// Uniform chunk size — used by unit tests and legacy call sites.
|
||||
public init(
|
||||
maxChunkDurationSeconds: TimeInterval,
|
||||
overlapDurationSeconds: TimeInterval,
|
||||
pauseExtensionMaxSeconds: TimeInterval,
|
||||
pauseRMSThreshold: Float,
|
||||
sampleRate: Int
|
||||
) {
|
||||
self.firstChunkDurationSeconds = maxChunkDurationSeconds
|
||||
self.subsequentChunkDurationSeconds = maxChunkDurationSeconds
|
||||
self.overlapDurationSeconds = overlapDurationSeconds
|
||||
self.pauseExtensionMaxSeconds = pauseExtensionMaxSeconds
|
||||
self.pauseRMSThreshold = pauseRMSThreshold
|
||||
self.sampleRate = sampleRate
|
||||
}
|
||||
|
||||
/// Backward-compatible alias for tests that read `maxChunkSamples`.
|
||||
public var maxChunkDurationSeconds: TimeInterval {
|
||||
subsequentChunkDurationSeconds
|
||||
}
|
||||
|
||||
public func maxChunkDurationSeconds(forChunkIndex index: Int) -> TimeInterval {
|
||||
index == 0 ? firstChunkDurationSeconds : subsequentChunkDurationSeconds
|
||||
}
|
||||
|
||||
public func maxChunkSamples(forChunkIndex index: Int) -> Int {
|
||||
Int(maxChunkDurationSeconds(forChunkIndex: index) * Double(sampleRate))
|
||||
}
|
||||
|
||||
public var maxChunkSamples: Int {
|
||||
Int(maxChunkDurationSeconds * Double(sampleRate))
|
||||
maxChunkSamples(forChunkIndex: 1)
|
||||
}
|
||||
|
||||
public var overlapSamples: Int {
|
||||
@@ -44,7 +77,8 @@ public struct FlowUtteranceChunkConfig: Sendable, Equatable {
|
||||
|
||||
/// Default for keyboard Flow utterances (≤ 3 min, pipelined ASR).
|
||||
public static let flowDefault = FlowUtteranceChunkConfig(
|
||||
maxChunkDurationSeconds: 30,
|
||||
firstChunkDurationSeconds: 2.5,
|
||||
subsequentChunkDurationSeconds: 5.0,
|
||||
overlapDurationSeconds: 0.5,
|
||||
pauseExtensionMaxSeconds: 2,
|
||||
pauseRMSThreshold: 0.015,
|
||||
|
||||
@@ -82,6 +82,14 @@ public struct LLMProvider: Identifiable, Codable, Hashable, Sendable {
|
||||
apiKeyURL: URL(string: "https://platform.moonshot.cn/console/api-keys"),
|
||||
blurb: "Kimi · 长上下文 · Long context"
|
||||
),
|
||||
.init(
|
||||
id: "mimo",
|
||||
name: "小米 MiMo",
|
||||
defaultBaseURL: "https://api.xiaomimimo.com/v1",
|
||||
defaultModel: "mimo-v2.5",
|
||||
apiKeyURL: URL(string: "https://platform.xiaomimimo.com"),
|
||||
blurb: "mimo-v2.5 · 中文优化 · Chinese-optimized"
|
||||
),
|
||||
.init(
|
||||
id: "custom",
|
||||
name: "Custom · 自定义",
|
||||
|
||||
@@ -1,50 +0,0 @@
|
||||
// LocalASRBackend.swift
|
||||
// OSGKeyboard · Shared
|
||||
//
|
||||
// Identifies which on-device speech recognition engine to use when the
|
||||
// user picks the "local" engine (no cloud LLM polish). The shared
|
||||
// factory `ASRServiceFactory` dispatches on this enum; the settings UI
|
||||
// renders it as a picker.
|
||||
//
|
||||
// As of v0.2.0 the only on-device backend is iOS 26 `SpeechAnalyzer`
|
||||
// + `DictationTranscriber`. The previous Qwen3-CoreML backend has
|
||||
// been removed: that path required a ~1.6 GB CoreML bundle, a local
|
||||
// SPM fork that pulled in mlx-swift, and significant app-side state
|
||||
// (download manager, warm-up service, model registry). We now keep the
|
||||
// local engine narrow — same iOS ASR the cloud engine already uses —
|
||||
// and let users opt into a cloud polish step after the transcript is
|
||||
// produced if they need stronger accuracy on noisy audio or dialectal
|
||||
// Chinese. See `LocalPolishConfig` for the post-ASR polish toggle.
|
||||
//
|
||||
// Why an enum in `Shared` rather than a `Bool`: the value must remain
|
||||
// serialisable into the App Group store (so the keyboard extension can
|
||||
// observe the selection) and exposed via `ProviderConfig` (UI binding).
|
||||
// Keeping the type stable even with a single case avoids a migration
|
||||
// the next time someone adds a non-cloud backend (e.g. whisper.cpp).
|
||||
|
||||
import Foundation
|
||||
|
||||
public enum LocalASRBackend: String, CaseIterable, Identifiable, Sendable, Codable {
|
||||
/// iOS 26 `SpeechAnalyzer` + `DictationTranscriber`. Always
|
||||
/// on-device, no asset download, ships with iOS. The only local
|
||||
/// backend in v0.2.0.
|
||||
case speechAnalyzer
|
||||
|
||||
public var id: String { rawValue }
|
||||
|
||||
/// Localisation key for the human label in the settings picker.
|
||||
public var labelKey: String {
|
||||
"asr.backend.speechAnalyzer.label"
|
||||
}
|
||||
|
||||
/// Localisation key for the one-line subtitle shown under the label.
|
||||
public var blurbKey: String {
|
||||
"asr.backend.speechAnalyzer.blurb"
|
||||
}
|
||||
|
||||
/// Whether this backend needs the user to download a model file
|
||||
/// before it can run. Always `false` for iOS-bundled speech.
|
||||
public var requiresModelDownload: Bool {
|
||||
false
|
||||
}
|
||||
}
|
||||
@@ -108,7 +108,9 @@ extension PersonalDictionary.Entry {
|
||||
/// learner. Users can re-classify later from Settings.
|
||||
public static func inferCategory(for term: String) -> Category {
|
||||
let hasUpper = term.contains(where: { $0.isUppercase })
|
||||
let hasDigit = term.contains(where: { $0.isNumber })
|
||||
let hasDigit = term.unicodeScalars.contains { scalar in
|
||||
CharacterSet.decimalDigits.contains(scalar) && scalar.isASCII
|
||||
}
|
||||
let hasLatin = term.unicodeScalars.contains { scalar in
|
||||
CharacterSet.letters.contains(scalar) && scalar.isASCII
|
||||
}
|
||||
|
||||
@@ -15,144 +15,108 @@ import Combine
|
||||
public final class ProviderConfig: ObservableObject, @unchecked Sendable {
|
||||
public static let shared = ProviderConfig()
|
||||
|
||||
private enum Key {
|
||||
static let providerId = "config.providerId"
|
||||
static let baseURL = "config.baseURL"
|
||||
// Legacy: apiKey used to live in UserDefaults before the
|
||||
// migration. We still read it once (see init below) and then
|
||||
// delete the entry, but no other code path touches this key.
|
||||
static let apiKeyLegacy = "config.apiKey"
|
||||
static let model = "config.model"
|
||||
static let modeId = "config.modeId"
|
||||
static let localeId = "config.localeId"
|
||||
static let engineMode = "config.engineMode"
|
||||
static let hasCompletedOnboarding = "config.hasCompletedOnboarding"
|
||||
static let onboardingPage = "config.onboardingPage"
|
||||
static let hasAcknowledgedCloudSharing = "config.hasAcknowledgedCloudSharing"
|
||||
// Which on-device ASR engine to use when `engineMode == "local"`.
|
||||
// Persisted in the App Group so the keyboard can read the
|
||||
// selection even though it never instantiates the backend itself.
|
||||
static let localASRBackend = "config.localASRBackend"
|
||||
static let uiLanguage = "config.uiLanguage"
|
||||
// v0.2.0: optional cloud polish step after on-device ASR finishes
|
||||
// in the local engine. Default `false` — keeps the local engine
|
||||
// truly local unless the user explicitly opts in.
|
||||
static let localModeCloudPolishEnabled = "config.localModeCloudPolishEnabled"
|
||||
// v0.2.1: optional translation step after ASR. The
|
||||
// post-ASR transcript is routed through the same LLM with a
|
||||
// translate-and-polish prompt targeting `translationTargetLocaleId`.
|
||||
// Mutually exclusive with the local-only promise — see `TranslationPolicy`.
|
||||
//
|
||||
// v0.2.1 follow-up: `config.translationEnabled` was *removed*
|
||||
// as a persisted key — translation is now derived from
|
||||
// `translationTargetLocaleId` (== offLocaleId means "off"). The
|
||||
// store still tolerates legacy reads of the old key so users
|
||||
// who upgraded from a build that wrote it don't see a flash of
|
||||
// "on" state during init, but new writes never touch the key.
|
||||
static let translationTargetLocaleId = "config.translationTargetLocaleId"
|
||||
static let handednessPreference = "config.handednessPreference"
|
||||
static let cursorDragNavigationEnabled = "config.cursorDragNavigationEnabled"
|
||||
// v0.3.0: how aggressively the LLM should rewrite transcripts.
|
||||
static let polishIntensity = "config.polishIntensity"
|
||||
}
|
||||
|
||||
@Published public var providerId: String {
|
||||
didSet {
|
||||
defaults.set(providerId, forKey: Key.providerId)
|
||||
// Keep API keys isolated per provider: switching provider in
|
||||
// Settings loads that provider's key instead of reusing the
|
||||
// previously selected vendor's key.
|
||||
guard !isApplyingConfiguration, providerId != configuration.providerId else { return }
|
||||
configuration.providerId = providerId
|
||||
isSyncingProviderAPIKey = true
|
||||
apiKey = Keychain.apiKey(for: providerId) ?? ""
|
||||
apiKey = configuration.apiKey
|
||||
isSyncingProviderAPIKey = false
|
||||
persistConfiguration()
|
||||
}
|
||||
}
|
||||
@Published public var baseURL: String {
|
||||
didSet { defaults.set(baseURL, forKey: Key.baseURL) }
|
||||
didSet {
|
||||
guard !isApplyingConfiguration, baseURL != configuration.baseURL else { return }
|
||||
configuration.baseURL = baseURL
|
||||
persistConfiguration()
|
||||
}
|
||||
}
|
||||
@Published public var apiKey: String {
|
||||
didSet {
|
||||
// Skip the round-trip on init — we read from Keychain and
|
||||
// writing the same value back is wasteful.
|
||||
guard oldValue != apiKey, !isSyncingProviderAPIKey else { return }
|
||||
do {
|
||||
try Keychain.setAPIKey(apiKey, for: providerId)
|
||||
} catch {
|
||||
#if DEBUG
|
||||
print("⚠️ [OSGKeyboard] Keychain write failed: \(error)")
|
||||
#endif
|
||||
OSGLog.config.warning("Keychain write failed: \(error.localizedDescription, privacy: .public)")
|
||||
}
|
||||
}
|
||||
}
|
||||
@Published public var model: String {
|
||||
didSet { defaults.set(model, forKey: Key.model) }
|
||||
didSet {
|
||||
guard !isApplyingConfiguration, model != configuration.model else { return }
|
||||
configuration.model = model
|
||||
persistConfiguration()
|
||||
}
|
||||
}
|
||||
@Published public var modeId: String {
|
||||
didSet { defaults.set(modeId, forKey: Key.modeId) }
|
||||
didSet {
|
||||
guard !isApplyingConfiguration, modeId != configuration.modeId else { return }
|
||||
configuration.modeId = modeId
|
||||
persistConfiguration()
|
||||
}
|
||||
}
|
||||
@Published public var localeId: String {
|
||||
didSet { defaults.set(localeId, forKey: Key.localeId) }
|
||||
didSet {
|
||||
guard !isApplyingConfiguration, localeId != configuration.localeId else { return }
|
||||
configuration.localeId = localeId
|
||||
persistConfiguration()
|
||||
}
|
||||
}
|
||||
/// "local" → on-device ASR + built-in DeepSeek polish.
|
||||
/// "cloud" → on-device ASR + user's cloud LLM polish.
|
||||
@Published public var engineMode: String {
|
||||
didSet {
|
||||
defaults.set(engineMode, forKey: Key.engineMode)
|
||||
guard !isApplyingConfiguration, engineMode != configuration.engineMode else { return }
|
||||
configuration.engineMode = engineMode
|
||||
applyEngineModeSideEffects()
|
||||
persistConfiguration()
|
||||
}
|
||||
}
|
||||
@Published public var hasCompletedOnboarding: Bool {
|
||||
didSet {
|
||||
defaults.set(hasCompletedOnboarding, forKey: Key.hasCompletedOnboarding)
|
||||
guard !isApplyingConfiguration,
|
||||
hasCompletedOnboarding != configuration.hasCompletedOnboarding else { return }
|
||||
configuration.hasCompletedOnboarding = hasCompletedOnboarding
|
||||
if hasCompletedOnboarding {
|
||||
configuration.onboardingPage = 0
|
||||
onboardingPage = 0
|
||||
}
|
||||
persistConfiguration()
|
||||
}
|
||||
}
|
||||
/// Persisted onboarding step so returning from Settings does not reset progress.
|
||||
@Published public var onboardingPage: Int {
|
||||
didSet { defaults.set(onboardingPage, forKey: Key.onboardingPage) }
|
||||
didSet {
|
||||
guard !isApplyingConfiguration, onboardingPage != configuration.onboardingPage else { return }
|
||||
configuration.onboardingPage = onboardingPage
|
||||
persistConfiguration()
|
||||
}
|
||||
}
|
||||
/// User confirmed that Cloud polish sends transcripts to their configured third-party API.
|
||||
@Published public var hasAcknowledgedCloudSharing: Bool {
|
||||
didSet { defaults.set(hasAcknowledgedCloudSharing, forKey: Key.hasAcknowledgedCloudSharing) }
|
||||
}
|
||||
/// Which on-device ASR engine backs the "local" engine mode. Only
|
||||
/// consulted when `isLocalEngine == true`; the cloud engine always
|
||||
/// uses `SpeechAnalyzer`.
|
||||
@Published public var localASRBackend: LocalASRBackend {
|
||||
didSet { defaults.set(localASRBackend.rawValue, forKey: Key.localASRBackend) }
|
||||
}
|
||||
/// When `engineMode == "local"`, optionally route the ASR transcript
|
||||
/// through the user's configured LLM (DeepSeek by default) before
|
||||
/// inserting at the cursor. The polish step runs through the same
|
||||
/// `LLMClient` + `PolishingService` stack the cloud engine uses.
|
||||
///
|
||||
/// Defaults to `false` — the local engine is ASR-only out of the
|
||||
/// box. Users opt in from Settings when the iOS ASR output isn't
|
||||
/// strong enough (noisy far-field audio, dialectal Chinese, etc.).
|
||||
@Published public var localModeCloudPolishEnabled: Bool {
|
||||
didSet {
|
||||
defaults.set(localModeCloudPolishEnabled, forKey: Key.localModeCloudPolishEnabled)
|
||||
AppGroupConfigDarwin.postConfigChanged()
|
||||
guard !isApplyingConfiguration,
|
||||
hasAcknowledgedCloudSharing != configuration.hasAcknowledgedCloudSharing else { return }
|
||||
configuration.hasAcknowledgedCloudSharing = hasAcknowledgedCloudSharing
|
||||
persistConfiguration()
|
||||
}
|
||||
}
|
||||
/// Host-app UI language. Also mirrored to the App Group for the keyboard extension.
|
||||
@Published public var uiLanguage: AppUILanguage {
|
||||
didSet { defaults.set(uiLanguage.rawValue, forKey: Key.uiLanguage) }
|
||||
didSet {
|
||||
guard !isApplyingConfiguration, uiLanguage != configuration.uiLanguage else { return }
|
||||
configuration.uiLanguage = uiLanguage
|
||||
persistConfiguration()
|
||||
}
|
||||
}
|
||||
/// v0.2.1: whether to translate the transcript into
|
||||
/// `translationTargetLocaleId` before insertion. **Derived** —
|
||||
/// translation is on iff the user has selected a target locale
|
||||
/// (i.e. the persisted id is anything other than
|
||||
/// `TranslationLanguageCatalog.offLocaleId`). Default off.
|
||||
///
|
||||
/// This used to be a stored `@Published var ... { didSet }` but the
|
||||
/// chip / picker now writes the locale directly; collapsing the
|
||||
/// pair into one field removes the "two writes out of sync" bug
|
||||
/// surface entirely.
|
||||
public var translationEnabled: Bool {
|
||||
translationTargetLocaleId != TranslationLanguageCatalog.offLocaleId
|
||||
configuration.translationEnabled
|
||||
}
|
||||
/// v0.2.1: BCP-47-ish target language id (e.g. `en`, `ja`, `ko`) the
|
||||
/// translate-and-polish prompt should produce. Default `"off"` —
|
||||
@@ -161,31 +125,37 @@ public final class ProviderConfig: ObservableObject, @unchecked Sendable {
|
||||
/// the user's choice without a host-app round-trip).
|
||||
@Published public var translationTargetLocaleId: String {
|
||||
didSet {
|
||||
defaults.set(translationTargetLocaleId, forKey: Key.translationTargetLocaleId)
|
||||
AppGroupConfigDarwin.postConfigChanged()
|
||||
guard !isApplyingConfiguration,
|
||||
translationTargetLocaleId != configuration.translationTargetLocaleId else { return }
|
||||
configuration.translationTargetLocaleId = translationTargetLocaleId
|
||||
persistConfiguration(postConfigChanged: true)
|
||||
}
|
||||
}
|
||||
/// Which hand the user holds the phone with — mirrors to the keyboard
|
||||
/// extension so delete / return can swap on the bottom row.
|
||||
@Published public var handednessPreference: HandednessPreference {
|
||||
didSet {
|
||||
defaults.set(handednessPreference.rawValue, forKey: Key.handednessPreference)
|
||||
AppGroupConfigDarwin.postConfigChanged()
|
||||
guard !isApplyingConfiguration,
|
||||
handednessPreference != configuration.handednessPreference else { return }
|
||||
configuration.handednessPreference = handednessPreference
|
||||
persistConfiguration(postConfigChanged: true)
|
||||
}
|
||||
}
|
||||
|
||||
/// Press-and-drag pads beside the mic for four-way caret movement.
|
||||
@Published public var cursorDragNavigationEnabled: Bool {
|
||||
didSet {
|
||||
defaults.set(cursorDragNavigationEnabled, forKey: Key.cursorDragNavigationEnabled)
|
||||
AppGroupConfigDarwin.postConfigChanged()
|
||||
guard !isApplyingConfiguration,
|
||||
cursorDragNavigationEnabled != configuration.cursorDragNavigationEnabled else { return }
|
||||
configuration.cursorDragNavigationEnabled = cursorDragNavigationEnabled
|
||||
persistConfiguration(postConfigChanged: true)
|
||||
}
|
||||
}
|
||||
|
||||
/// Whether the pipeline should run translate-and-polish (not just
|
||||
/// polish). Both engines honour the selected target locale.
|
||||
public var isTranslationEffective: Bool {
|
||||
translationEnabled
|
||||
configuration.isTranslationEffective
|
||||
}
|
||||
|
||||
/// Translation picker visibility — available on both engines.
|
||||
@@ -194,21 +164,22 @@ public final class ProviderConfig: ObservableObject, @unchecked Sendable {
|
||||
/// v0.3.0: how aggressively the LLM should rewrite the ASR
|
||||
/// transcript. Default is `medium` (Typeless-equivalent).
|
||||
@Published public var polishIntensity: PolishIntensity {
|
||||
didSet { defaults.set(polishIntensity.rawValue, forKey: Key.polishIntensity) }
|
||||
didSet {
|
||||
guard !isApplyingConfiguration, polishIntensity != configuration.polishIntensity else { return }
|
||||
configuration.polishIntensity = polishIntensity
|
||||
persistConfiguration()
|
||||
}
|
||||
}
|
||||
|
||||
public var isConfigured: Bool {
|
||||
// Local engine (on-device ASR only) doesn't need an API key,
|
||||
// base URL, or model — the LLM round-trip is skipped entirely.
|
||||
// Treat it as always-configured so onboarding's "Next" button
|
||||
// enables the moment the user picks the local path, instead
|
||||
// of forcing them to fill in cloud fields they won't use.
|
||||
// Local engine uses on-device ASR + built-in DeepSeek polish and
|
||||
// does not need a user API key. Cloud needs base URL, key, and model.
|
||||
if isLocalEngine { return true }
|
||||
return !baseURL.isEmpty && !apiKey.isEmpty && !model.isEmpty
|
||||
}
|
||||
|
||||
/// On-device ASR only; no cloud API required.
|
||||
public var isLocalEngine: Bool { engineMode == "local" }
|
||||
public var isLocalEngine: Bool { configuration.isLocalEngine }
|
||||
|
||||
/// Local engine always polishes via the built-in DeepSeek path.
|
||||
public var shouldPolishLocalTranscript: Bool { isLocalEngine }
|
||||
@@ -217,83 +188,37 @@ public final class ProviderConfig: ObservableObject, @unchecked Sendable {
|
||||
public var localModeProviderId: String { "deepseek" }
|
||||
|
||||
private let defaults: UserDefaults
|
||||
private var configuration: AppGroupConfiguration
|
||||
private var isApplyingConfiguration = false
|
||||
private var isSyncingProviderAPIKey = false
|
||||
|
||||
public init(defaults: UserDefaults? = nil) {
|
||||
let resolvedDefaults: UserDefaults = defaults ?? (AppGroup.isAvailable ? AppGroup.defaults : .standard)
|
||||
guard let resolvedDefaults = defaults ?? AppGroup.defaultsIfAvailable else {
|
||||
preconditionFailure(
|
||||
"ProviderConfig requires App Group or injected UserDefaults — " +
|
||||
"check AppGroup.isAvailable before constructing."
|
||||
)
|
||||
}
|
||||
self.defaults = resolvedDefaults
|
||||
let pid = resolvedDefaults.string(forKey: Key.providerId) ?? "openai"
|
||||
let preset = LLMProvider.provider(id: pid)
|
||||
self.providerId = pid
|
||||
self.baseURL = resolvedDefaults.string(forKey: Key.baseURL) ?? preset.defaultBaseURL
|
||||
self.configuration = AppGroupConfiguration.load(fromAvailable: resolvedDefaults)
|
||||
|
||||
// Resolve the API key with a one-shot migration from the legacy
|
||||
// UserDefaults slot. After this runs once, `Key.apiKeyLegacy`
|
||||
// is empty in the suite and all subsequent reads go through the
|
||||
// Keychain.
|
||||
self.apiKey = ProviderConfig.resolveAPIKey(defaults: resolvedDefaults, providerId: pid)
|
||||
|
||||
self.model = resolvedDefaults.string(forKey: Key.model) ?? preset.defaultModel
|
||||
self.modeId = resolvedDefaults.string(forKey: Key.modeId) ?? "polish"
|
||||
self.localeId = resolvedDefaults.string(forKey: Key.localeId) ?? "auto"
|
||||
self.engineMode = resolvedDefaults.string(forKey: Key.engineMode) ?? "cloud"
|
||||
self.hasCompletedOnboarding = resolvedDefaults.bool(forKey: Key.hasCompletedOnboarding)
|
||||
let savedPage = resolvedDefaults.integer(forKey: Key.onboardingPage)
|
||||
self.onboardingPage = savedPage > 0 ? savedPage : 0
|
||||
self.hasAcknowledgedCloudSharing = resolvedDefaults.bool(forKey: Key.hasAcknowledgedCloudSharing)
|
||||
// Tolerate missing / unknown raw values (e.g. an enum case that
|
||||
// was renamed in a later build) by falling back to the default
|
||||
// rather than crashing inside `RawRepresentable.init`.
|
||||
let rawBackend = resolvedDefaults.string(forKey: Key.localASRBackend) ?? LocalASRBackend.speechAnalyzer.rawValue
|
||||
self.localASRBackend = LocalASRBackend(rawValue: rawBackend) ?? .speechAnalyzer
|
||||
// v0.2.0: local-mode cloud polish toggle. Defaults off; users
|
||||
// opt in from Settings when iOS ASR is too lossy for their
|
||||
// environment. `object(forKey:) == nil` covers fresh installs
|
||||
// and upgrades from builds that never wrote the key.
|
||||
if resolvedDefaults.object(forKey: Key.localModeCloudPolishEnabled) == nil {
|
||||
self.localModeCloudPolishEnabled = false
|
||||
} else {
|
||||
self.localModeCloudPolishEnabled = resolvedDefaults.bool(forKey: Key.localModeCloudPolishEnabled)
|
||||
}
|
||||
self.uiLanguage = AppUILanguage.fromStored(
|
||||
resolvedDefaults.string(forKey: Key.uiLanguage)
|
||||
)
|
||||
// v0.2.1 follow-up: `translationEnabled` is now derived from
|
||||
// `translationTargetLocaleId` — no separate init read.
|
||||
// Default the locale id to `offLocaleId` so existing installs
|
||||
// that never picked a target language stay in the "off" state
|
||||
// (the previous build's default of `"en"` would silently turn
|
||||
// translation on for every upgraded user; off is the safe
|
||||
// conservative default that matches the picker / chip UX).
|
||||
self.translationTargetLocaleId = resolvedDefaults.string(forKey: Key.translationTargetLocaleId)
|
||||
?? TranslationLanguageCatalog.offLocaleId
|
||||
self.handednessPreference = HandednessPreference.fromStored(
|
||||
resolvedDefaults.string(forKey: Key.handednessPreference)
|
||||
)
|
||||
if resolvedDefaults.object(forKey: Key.cursorDragNavigationEnabled) == nil {
|
||||
self.cursorDragNavigationEnabled = true
|
||||
} else {
|
||||
self.cursorDragNavigationEnabled = resolvedDefaults.bool(forKey: Key.cursorDragNavigationEnabled)
|
||||
}
|
||||
// v0.3.0: polish intensity. Default to `.medium` for new
|
||||
// installs; legacy `"off"` migrates to `.medium`.
|
||||
if let raw = resolvedDefaults.string(forKey: Key.polishIntensity) {
|
||||
self.polishIntensity = PolishIntensity.resolve(storedRawValue: raw)
|
||||
if raw == PolishIntensity.legacyOffRawValue {
|
||||
resolvedDefaults.set(PolishIntensity.medium.rawValue, forKey: Key.polishIntensity)
|
||||
}
|
||||
} else {
|
||||
self.polishIntensity = .default
|
||||
}
|
||||
|
||||
// Cloud no longer exposes off/transcribe; migrate legacy values.
|
||||
if self.engineMode == "cloud", self.modeId != "polish" {
|
||||
self.modeId = "polish"
|
||||
}
|
||||
// DeepSeek is local-engine only — never a cloud picker choice.
|
||||
if self.engineMode == "cloud", self.providerId == "deepseek" {
|
||||
apply(preset: LLMProvider.provider(id: "openai"))
|
||||
}
|
||||
isApplyingConfiguration = true
|
||||
providerId = configuration.providerId
|
||||
baseURL = configuration.baseURL
|
||||
apiKey = configuration.apiKey
|
||||
model = configuration.model
|
||||
modeId = configuration.modeId
|
||||
localeId = configuration.localeId
|
||||
engineMode = configuration.engineMode
|
||||
hasCompletedOnboarding = configuration.hasCompletedOnboarding
|
||||
onboardingPage = configuration.onboardingPage
|
||||
hasAcknowledgedCloudSharing = configuration.hasAcknowledgedCloudSharing
|
||||
uiLanguage = configuration.uiLanguage
|
||||
translationTargetLocaleId = configuration.translationTargetLocaleId
|
||||
handednessPreference = configuration.handednessPreference
|
||||
cursorDragNavigationEnabled = configuration.cursorDragNavigationEnabled
|
||||
polishIntensity = configuration.polishIntensity
|
||||
isApplyingConfiguration = false
|
||||
}
|
||||
|
||||
/// Keep cloud vs local provider choices isolated when the user
|
||||
@@ -304,29 +229,15 @@ public final class ProviderConfig: ObservableObject, @unchecked Sendable {
|
||||
}
|
||||
}
|
||||
|
||||
/// Read the API key from the Keychain, falling back to a one-time
|
||||
/// migration from the legacy UserDefaults slot.
|
||||
private static func resolveAPIKey(defaults: UserDefaults, providerId: String) -> String {
|
||||
if let stored = Keychain.apiKey(for: providerId), !stored.isEmpty {
|
||||
return stored
|
||||
private func persistConfiguration(postConfigChanged: Bool = false) {
|
||||
configuration.save(to: defaults)
|
||||
if postConfigChanged {
|
||||
AppGroupConfigDarwin.postConfigChanged()
|
||||
}
|
||||
// Migration path: old builds stored one global key under
|
||||
// Keychain account "current". Move it to the active provider.
|
||||
if let legacyKeychain = Keychain.legacyAPIKey(), !legacyKeychain.isEmpty {
|
||||
try? Keychain.setAPIKey(legacyKeychain, for: providerId)
|
||||
try? Keychain.deleteLegacyAPIKey()
|
||||
return legacyKeychain
|
||||
}
|
||||
if let legacy = defaults.string(forKey: Key.apiKeyLegacy),
|
||||
!legacy.isEmpty {
|
||||
try? Keychain.setAPIKey(legacy, for: providerId)
|
||||
defaults.removeObject(forKey: Key.apiKeyLegacy)
|
||||
return legacy
|
||||
}
|
||||
return ""
|
||||
}
|
||||
|
||||
public func apply(preset: LLMProvider) {
|
||||
isApplyingConfiguration = true
|
||||
providerId = preset.id
|
||||
if !preset.defaultBaseURL.isEmpty {
|
||||
baseURL = preset.defaultBaseURL
|
||||
@@ -334,15 +245,31 @@ public final class ProviderConfig: ObservableObject, @unchecked Sendable {
|
||||
if !preset.defaultModel.isEmpty {
|
||||
model = preset.defaultModel
|
||||
}
|
||||
configuration.providerId = providerId
|
||||
configuration.baseURL = baseURL
|
||||
configuration.model = model
|
||||
isSyncingProviderAPIKey = true
|
||||
apiKey = configuration.apiKey
|
||||
isSyncingProviderAPIKey = false
|
||||
isApplyingConfiguration = false
|
||||
persistConfiguration()
|
||||
}
|
||||
|
||||
public func reset() {
|
||||
providerId = "openai"
|
||||
isApplyingConfiguration = true
|
||||
let preset = LLMProvider.provider(id: "openai")
|
||||
providerId = preset.id
|
||||
baseURL = preset.defaultBaseURL
|
||||
apiKey = ""
|
||||
model = preset.defaultModel
|
||||
handednessPreference = .left
|
||||
hasAcknowledgedCloudSharing = false
|
||||
configuration.providerId = preset.id
|
||||
configuration.baseURL = preset.defaultBaseURL
|
||||
configuration.model = preset.defaultModel
|
||||
configuration.handednessPreference = .left
|
||||
configuration.hasAcknowledgedCloudSharing = false
|
||||
isApplyingConfiguration = false
|
||||
persistConfiguration()
|
||||
}
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user