mirror of
https://github.com/anomalyco/hex.git
synced 2026-11-22 21:23:02 +01:00
* iOS keyboard: live words over the keys while dictating (stage 1, two looks)
What: while the keyboard records (in-keyboard dictation, iOS 26+), the capture's
16-bit chunks also feed Apple's SpeechAnalyzer (SpeechTranscriber with volatile and
fast results, DictationTranscriber's progressive preset for its languages), and the
words so far show over the keys: settled words in the keys' ink, the tentative tail
muted, newest line in view, whole lines at the top. Two looks behind a temporary App
Group key (debug.liveDictationLook): dim (keys at 10%) and hide. The preview takes every
touch on the keys, so nothing types while it shows; the recording bar, its cancel and
done, and the globe (drawn over the words where the keyboard has its own) work as before.
The final text still comes from the recording exactly as today; the live recogniser is
stopped when the capture stops (model lingering for the final pass).
Why: Ryan asked for the dictation to show in real time over a hidden or dimmed keyboard.
LiveTranscriptSource is one small protocol so the source can be swapped:
ParakeetLiveTranscriber (the audio so far again about once a second, a 20 s window) is
written but not wired. Never worse than today: no preview when neither switch wants it,
when less than 24 MB is left, when the language's model isn't on the iPhone (no
download), on an error, or before iOS 26. Hooks in KeyboardLocalDictation stay in the
capture's events and phase, away from the transcription call.
Verified: Studio simulator (iPhone 17 Pro iOS 27.2, SE iOS 26.5), real keyboard in a
host app with results recorded from Apple's recogniser on macOS 27.2.1 replayed (the
simulator has no speech models; LiveTranscriptReplay is simulator-only): both looks,
light and dark, upright and turned, a 60 s dictation scrolling, taps on Q and space
while recording type nothing with the preview (and "Q " without it), the globe switches
keyboards on the SE, no model -> no preview and the keys as today. Measured on macOS
with the same source: first words 0.5-1.0 s after the first word ends, word lag median
0.56-0.76 s, 2.1% of shown words rewritten, +0.4-2.6 MB in the feeding process.
code-shape.py passes. No settings UI yet (stage 2).
* iOS keyboard: Show what I say (hide look) and Type as I speak
What: look (b) only. While the keyboard records by itself the keys go and the words
so far take their place under the recording bar; the dim look and its temporary
debug.liveDictationLook key are gone. Two switches in Settings › Dictation on their
own card under the style card, iOS 26 and later only, Simplified Chinese:
- Show what I say (on by default): the preview.
- Type as I speak (off by default, "Words can change until you stop"):
LiveFieldTyping types the live words through the text document proxy, retyping only
the changed tail, and keeps exactly what it typed. It edits only while the text
before the cursor still ends with that (smart quotes allowed for; a host showing only
the end must show the part that changes); otherwise it stops and leaves the field,
and the dictation goes in at the cursor as today. When the dictation's text goes in,
KeyboardLiveTyping.handOver takes the words out first and returns the field as it was,
so spacing, Undo dictation (the whole dictation), Learn how I type and typing logs see
exactly what they did before. Cancel, a failure, nothing heard and the keyboard hiding
take the words out (a held dictation goes in at the next keyboard as before). Never
in secure, password, one-time code or card number fields, over a selection or a
two-thumb Dictate's range, or with Translate what I say on (Ryan's call).
- Never worse than today: live words only with 32 MB left once the capture runs (the
speech model's 24 MB transcription headroom plus the recogniser's share), stopped under
24 MB or on a memory warning; the preview never shows before a recogniser is running,
so a dictation without one never flashes it.
- The hand-over: the preview fades with the words it had while the keys and the
dictation's own text come in; with Parakeet now English's default the field's words
can differ slightly from the preview's (Apple's), and never flash in the preview.
Why: Ryan picked look (b) and the two switch names at the checkpoint.
Verified: Studio simulator (iPhone 17 Pro, iOS 27.2) with the replayed recogniser
results: the field typed and revised as the words came ("Quick note: Hey, I am running
about 10 minutes late. Can you grab us a table by...") and back to "Quick note:" after
Cancel; Settings › Dictation in English, dark with both on, and Simplified Chinese.
New tests in DictationLearningTests (LiveDictationTests: accumulator, joining, switch
defaults, typing and revising, stopping when the field changes or reports nothing,
private fields). README describes what the keyboard shows while recording.
code-shape.py passes.
* iOS keyboard: the preview draws only the end of a long dictation
What: LiveDictationPreviewView draws its words with one label in a clipping box, moved up by
whole lines once they overflow it, and only about the box's worth of text and three lines more
(from a word's start). It replaces the scrolling UITextView, so the words can no longer be
dragged back.
Why: on the iPhone 17 Pro a 60 s dictation with the preview held up to 27 MB more in the
keyboard (peak 103 MB, against 60 MB with the recogniser on and the preview off); the Studio
simulator showed the same, +24.5 MB.
Verified: Studio simulator, the 60 s replay: +24.5 MB (text view) -> +17.3 (capped text view)
-> +5.7 MB with the label, against +7.0 MB for the same recording without the preview. Test
for the tail in LiveDictationTests.
* iOS keyboard: the preview's tail keeps a word that starts right at the cut
Verified: LiveDictationTests on the Studio simulator.
* iOS keyboard: Type as I speak's words are replaced by the held text after the keyboard hides
What: when the keyboard hides mid-dictation, KeyboardLiveTyping notes the words it typed
(LiveDictationSettings.leftover, in the App Group) as well as taking them out. When the held
dictation goes in (inserted by itself or by Paste dictation), it replaces those words if the
text before the cursor still ends with them (LiveFieldTyping(resuming:in:)); otherwise it goes in
at the cursor as before. A new dictation clears the note.
Why: on the iPhone 17 Pro, pressing Home mid-dictation left the typed words in the field (an
app gone to the background takes no edits) and, back in the app (a new document identifier),
the strip offered Paste dictation, which would have put the dictation after them.
Verified: LiveDictationTests on the Studio simulator (9 tests).
* iOS README: what the keyboard shows while recording, measured on the iPhone 17 Pro
Adds the phone's numbers (first words about 1.2-1.3 s after speech starts; 60 s dictation 60.1 MB
with the preview against 58.1 MB without, peaks 66.5 against 60.1), the hidden-keyboard case for
Type as I speak, and what is verified where and what isn't (VoiceOver on a device, lagging hosts).
161 lines
6.4 KiB
Swift
161 lines
6.4 KiB
Swift
import Combine
|
||
import Foundation
|
||
|
||
/// The words heard so far while a dictation records, shown over the keys while it does:
|
||
/// `settled` words the recogniser won't change again, then the `tentative` tail it still may.
|
||
nonisolated struct LiveTranscript: Equatable, Sendable {
|
||
var settled = ""
|
||
var tentative = ""
|
||
|
||
init(settled: String = "", tentative: String = "") {
|
||
self.settled = settled
|
||
self.tentative = tentative
|
||
}
|
||
|
||
/// Both parts as one line, one space between them unless either already has one.
|
||
var text: String {
|
||
settled + tentativeJoined
|
||
}
|
||
|
||
/// The tentative words as they follow the settled ones.
|
||
var tentativeJoined: String {
|
||
guard let last = settled.last, let first = tentative.first,
|
||
!last.isWhitespace, !first.isWhitespace,
|
||
!LiveTranscript.joinsWithoutSpace(last, first) else { return tentative }
|
||
return " " + tentative
|
||
}
|
||
|
||
var isEmpty: Bool {
|
||
settled.allSatisfy(\.isWhitespace) && tentative.allSatisfy(\.isWhitespace)
|
||
}
|
||
|
||
/// Chinese and Japanese run on without spaces; a mark never takes one before it.
|
||
static func joinsWithoutSpace(_ last: Character, _ first: Character) -> Bool {
|
||
if first.isPunctuation { return true }
|
||
guard let a = last.unicodeScalars.first?.value, let b = first.unicodeScalars.first?.value else { return false }
|
||
func written(_ v: UInt32) -> Bool {
|
||
(0x3041...0x30FF).contains(v) || (0x3400...0x4DBF).contains(v)
|
||
|| (0x4E00...0x9FFF).contains(v) || (0xF900...0xFAFF).contains(v) || (0x3000...0x303F).contains(v)
|
||
}
|
||
return written(a) || written(b)
|
||
}
|
||
}
|
||
|
||
/// A recogniser's results as one transcript: a final result settles its words and the latest
|
||
/// volatile one is the tentative tail (Apple's `SpeechTranscriber` replaces its volatile
|
||
/// result each time and never revisits a final one).
|
||
nonisolated struct LiveTranscriptAccumulator {
|
||
private var finals: [String] = []
|
||
private(set) var transcript = LiveTranscript()
|
||
|
||
@discardableResult
|
||
mutating func apply(_ text: String, isFinal: Bool) -> LiveTranscript {
|
||
if isFinal {
|
||
finals.append(text)
|
||
transcript = LiveTranscript(settled: Self.joined(finals), tentative: "")
|
||
} else {
|
||
transcript.tentative = text.trimmingCharacters(in: .whitespacesAndNewlines)
|
||
}
|
||
return transcript
|
||
}
|
||
|
||
/// Final results back to back, as Apple writes them, trimmed at both ends.
|
||
static func joined(_ segments: [String]) -> String {
|
||
var text = ""
|
||
for segment in segments {
|
||
let piece = segment.trimmingCharacters(in: .whitespacesAndNewlines)
|
||
guard !piece.isEmpty else { continue }
|
||
text = LiveTranscript(settled: text, tentative: piece).text
|
||
}
|
||
return text
|
||
}
|
||
}
|
||
|
||
/// Where live words come from while the keyboard records: Apple's speech recognition fed as
|
||
/// the audio arrives today, the keyboard's speech model later. `append` takes the capture's
|
||
/// 16-bit mono chunks; `stop` frees it at once and reports nothing more.
|
||
@MainActor
|
||
protocol LiveTranscriptSource: AnyObject {
|
||
var onUpdate: ((LiveTranscript) -> Void)? { get set }
|
||
/// The source gave up (no model, no memory, an error): the preview goes and the dictation
|
||
/// goes on as without it.
|
||
var onFailure: ((String) -> Void)? { get set }
|
||
func append(_ pcm: Data)
|
||
func stop()
|
||
}
|
||
|
||
/// What the keyboard shows while it records: nil while no live source runs (the keys stay as
|
||
/// they are), else the words so far. The preview and Type as I speak read it; changing it
|
||
/// never refreshes the rest of the keyboard.
|
||
@MainActor
|
||
final class LiveDictationText: ObservableObject {
|
||
@Published private(set) var transcript: LiveTranscript?
|
||
/// Settings › Dictation's two switches as this dictation started.
|
||
private(set) var showsPreview = true
|
||
private(set) var typesIntoField = false
|
||
/// Counts dictations, so what follows the words can tell a new one from the last.
|
||
private(set) var dictation = 0
|
||
|
||
/// A source started: the preview shows, empty until the first words.
|
||
func begin(showsPreview: Bool, typesIntoField: Bool) {
|
||
self.showsPreview = showsPreview
|
||
self.typesIntoField = typesIntoField
|
||
dictation += 1
|
||
transcript = LiveTranscript()
|
||
}
|
||
|
||
func update(_ next: LiveTranscript) {
|
||
guard transcript != nil, next != transcript else { return }
|
||
transcript = next
|
||
}
|
||
|
||
func end() {
|
||
if transcript != nil { transcript = nil }
|
||
}
|
||
}
|
||
|
||
/// Settings › Dictation's two switches for what happens while you speak, in the App Group so
|
||
/// the keyboard reads them: Show what I say (the words over the keys, on by default) and
|
||
/// Type as I speak (the words typed into the field, off by default).
|
||
nonisolated struct LiveDictationSettings {
|
||
static let previewKey = "settings.liveDictationPreview"
|
||
static let typingKey = "settings.liveDictationTyping"
|
||
static let leftoverKey = "liveDictation.leftover"
|
||
|
||
private let defaults: UserDefaults?
|
||
|
||
init(defaults: UserDefaults? = UserDefaults(suiteName: KeyboardBridge.appGroup)) {
|
||
self.defaults = defaults
|
||
}
|
||
|
||
var showsPreview: Bool {
|
||
get { defaults?.object(forKey: Self.previewKey) as? Bool ?? true }
|
||
nonmutating set { defaults?.set(newValue, forKey: Self.previewKey) }
|
||
}
|
||
|
||
var typesIntoField: Bool {
|
||
get { defaults?.object(forKey: Self.typingKey) as? Bool ?? false }
|
||
nonmutating set { defaults?.set(newValue, forKey: Self.typingKey) }
|
||
}
|
||
|
||
/// Either switch needs live words while recording.
|
||
var wantsLiveWords: Bool { showsPreview || typesIntoField }
|
||
|
||
/// The words Type as I speak typed in a field whose keyboard hid before the dictation went
|
||
/// in (an app gone to the background takes no edits), and when: the held text replaces them
|
||
/// where the field still ends with them. A new dictation clears them.
|
||
var leftover: (text: String, at: TimeInterval)? {
|
||
get {
|
||
guard let stored = defaults?.dictionary(forKey: Self.leftoverKey),
|
||
let text = stored["text"] as? String, let at = stored["at"] as? Double else { return nil }
|
||
return (text, at)
|
||
}
|
||
nonmutating set {
|
||
if let newValue {
|
||
defaults?.set(["text": newValue.text, "at": newValue.at], forKey: Self.leftoverKey)
|
||
} else {
|
||
defaults?.removeObject(forKey: Self.leftoverKey)
|
||
}
|
||
}
|
||
}
|
||
}
|