Speaker corrections: rename / merge / reassign + voice learning
Native editor to fix speaker-ID errors after transcription (modeled on recap-relay's correction UX): rename a speaker in the legend, merge two speakers, or reassign an individual transcript line. Saving rewrites speakers.json, re-renders transcript.md + recap.html, and updates the voiceprint memory — so a correction compounds: naming an "Unknown" speaker teaches that voice for future calls. - SpeakerEditing (pure, tested): replaceSpeaker (rename = merge-onto-existing), reassign, netNameMap (compose ops), and remap (apply a name map to a recap's structured fields + whole-word free text, so summaries/extras update without re-LLM). - RecapEditModel (@MainActor): loads speakers.json (+ optional recap.json + cluster_fingerprints.json); on save writes the resolved speakers.json, re-renders, and reconciles voiceprints — merge keeps the survivor's print; rename/name-an-Unknown enrolls the cluster's fingerprint under the new name. - TranscriptEditorView (SwiftUI) + EditorWindow (AppKit window for the LSUIElement app); menu gains "Edit speakers". - Pipeline now persists cluster_fingerprints.json (every cluster incl. Unknown) and recap.json (RecapFile) so the editor can learn voices + re-render offline. - RecapModels made Codable; TranscriptAssembler exposes allFingerprints; VoiceprintStore gains enroll() + merge(). 52/52 XCTest (6 new, incl. a full rename→artifacts→voiceprint round-trip on disk).
This commit is contained in:
Grant Gilliam
@@ -0,0 +1,32 @@
|
||||
import AppKit
|
||||
import SwiftUI
|
||||
|
||||
/// Hosts the speaker-correction editor in a standalone resizable window. A
|
||||
/// menu-bar (LSUIElement) app has no normal window scene, so we open one via
|
||||
/// AppKit and activate the app so it comes to the front.
|
||||
@MainActor
|
||||
final class EditorWindow {
|
||||
static let shared = EditorWindow()
|
||||
private var window: NSWindow?
|
||||
|
||||
func show(model: RecapEditModel) {
|
||||
if let window {
|
||||
window.contentViewController = NSHostingController(rootView: TranscriptEditorView(model: model))
|
||||
window.title = model.title
|
||||
NSApp.activate(ignoringOtherApps: true)
|
||||
window.makeKeyAndOrderFront(nil)
|
||||
return
|
||||
}
|
||||
let w = NSWindow(
|
||||
contentRect: NSRect(x: 0, y: 0, width: 640, height: 560),
|
||||
styleMask: [.titled, .closable, .resizable, .miniaturizable],
|
||||
backing: .buffered, defer: false)
|
||||
w.title = model.title
|
||||
w.isReleasedWhenClosed = false
|
||||
w.center()
|
||||
w.contentViewController = NSHostingController(rootView: TranscriptEditorView(model: model))
|
||||
window = w
|
||||
NSApp.activate(ignoringOtherApps: true)
|
||||
w.makeKeyAndOrderFront(nil)
|
||||
}
|
||||
}
|
||||
@@ -94,6 +94,9 @@ struct MenuBarView: View {
|
||||
if let recap = session.recapURL {
|
||||
Button("Open recap") { NSWorkspace.shared.open(recap) }
|
||||
}
|
||||
if session.canEditSpeakers {
|
||||
Button("Edit speakers") { session.editLastSession() }
|
||||
}
|
||||
Spacer()
|
||||
}
|
||||
if !transcriptText.isEmpty {
|
||||
|
||||
@@ -0,0 +1,152 @@
|
||||
import SwiftUI
|
||||
|
||||
/// Speaker-correction editor: rename a speaker, merge two speakers, or reassign an
|
||||
/// individual transcript line to a different speaker. Saving rewrites speakers.json,
|
||||
/// re-renders the recap, and updates the voiceprint memory.
|
||||
struct TranscriptEditorView: View {
|
||||
@ObservedObject var model: RecapEditModel
|
||||
|
||||
@State private var renameTarget: String?
|
||||
@State private var renameText = ""
|
||||
@State private var reassignNewIndex: Int?
|
||||
@State private var newSpeakerText = ""
|
||||
|
||||
var body: some View {
|
||||
VStack(alignment: .leading, spacing: 0) {
|
||||
header
|
||||
Divider()
|
||||
transcript
|
||||
Divider()
|
||||
footer
|
||||
}
|
||||
.frame(minWidth: 560, minHeight: 480)
|
||||
.sheet(item: Binding(get: { renameTarget.map { IdentifiableString($0) } },
|
||||
set: { renameTarget = $0?.value })) { item in
|
||||
renameSheet(for: item.value)
|
||||
}
|
||||
.sheet(item: Binding(get: { reassignNewIndex.map { IdentifiableInt($0) } },
|
||||
set: { reassignNewIndex = $0?.value })) { item in
|
||||
newSpeakerSheet(for: item.value)
|
||||
}
|
||||
}
|
||||
|
||||
// MARK: - Header / legend
|
||||
|
||||
private var header: some View {
|
||||
VStack(alignment: .leading, spacing: 8) {
|
||||
Text(model.title).font(.headline)
|
||||
Text("Fix speaker names, merge duplicates, or reassign a line. Saving updates the recap and remembers any names you set.")
|
||||
.font(.caption).foregroundStyle(.secondary)
|
||||
ScrollView(.horizontal, showsIndicators: false) {
|
||||
HStack(spacing: 8) {
|
||||
ForEach(model.speakers, id: \.self) { name in
|
||||
Menu {
|
||||
Button("Rename…") { renameText = name; renameTarget = name }
|
||||
if model.speakers.count > 1 {
|
||||
Menu("Merge into") {
|
||||
ForEach(model.speakers.filter { $0 != name }, id: \.self) { other in
|
||||
Button(other) { model.merge(name, into: other) }
|
||||
}
|
||||
}
|
||||
}
|
||||
} label: {
|
||||
Text(name).font(.caption).padding(.horizontal, 8).padding(.vertical, 3)
|
||||
.background(Color.accentColor.opacity(0.18)).clipShape(Capsule())
|
||||
}
|
||||
.menuStyle(.borderlessButton).fixedSize()
|
||||
}
|
||||
}
|
||||
.padding(.vertical, 2)
|
||||
}
|
||||
}
|
||||
.padding(12)
|
||||
}
|
||||
|
||||
// MARK: - Transcript
|
||||
|
||||
private var transcript: some View {
|
||||
List {
|
||||
ForEach(Array(model.segments.enumerated()), id: \.offset) { idx, seg in
|
||||
HStack(alignment: .top, spacing: 10) {
|
||||
Text(RecapAnalyzer.mmss(seg.start))
|
||||
.font(.caption.monospacedDigit()).foregroundStyle(.secondary)
|
||||
.frame(width: 52, alignment: .trailing)
|
||||
Menu {
|
||||
ForEach(model.speakers, id: \.self) { name in
|
||||
Button(name) { model.reassign(idx, to: name) }
|
||||
}
|
||||
Divider()
|
||||
Button("New name…") { newSpeakerText = ""; reassignNewIndex = idx }
|
||||
} label: {
|
||||
Text(seg.speaker).font(.caption.weight(.semibold))
|
||||
.frame(width: 96, alignment: .leading)
|
||||
}
|
||||
.menuStyle(.borderlessButton).fixedSize()
|
||||
Text(seg.text ?? "").font(.callout).textSelection(.enabled)
|
||||
.frame(maxWidth: .infinity, alignment: .leading)
|
||||
}
|
||||
.padding(.vertical, 2)
|
||||
}
|
||||
}
|
||||
.listStyle(.inset)
|
||||
}
|
||||
|
||||
// MARK: - Footer
|
||||
|
||||
private var footer: some View {
|
||||
HStack {
|
||||
Button("Save corrections") { model.save() }
|
||||
.keyboardShortcut("s", modifiers: .command)
|
||||
.disabled(!model.dirty)
|
||||
if let s = model.status { Text(s).font(.caption).foregroundStyle(.green) }
|
||||
Spacer()
|
||||
if let recap = recapURL {
|
||||
Button("Open recap") { NSWorkspace.shared.open(recap) }
|
||||
}
|
||||
}
|
||||
.padding(12)
|
||||
}
|
||||
|
||||
private var recapURL: URL? {
|
||||
let u = model.folder.appendingPathComponent("recap.html")
|
||||
return FileManager.default.fileExists(atPath: u.path) ? u : nil
|
||||
}
|
||||
|
||||
// MARK: - Sheets
|
||||
|
||||
private func renameSheet(for name: String) -> some View {
|
||||
VStack(alignment: .leading, spacing: 12) {
|
||||
Text("Rename “\(name)”").font(.headline)
|
||||
TextField("New name", text: $renameText).textFieldStyle(.roundedBorder).frame(width: 260)
|
||||
HStack {
|
||||
Spacer()
|
||||
Button("Cancel") { renameTarget = nil }
|
||||
Button("Rename") {
|
||||
model.rename(name, to: renameText)
|
||||
renameTarget = nil
|
||||
}.keyboardShortcut(.defaultAction).disabled(renameText.trimmingCharacters(in: .whitespaces).isEmpty)
|
||||
}
|
||||
}
|
||||
.padding(16)
|
||||
}
|
||||
|
||||
private func newSpeakerSheet(for index: Int) -> some View {
|
||||
VStack(alignment: .leading, spacing: 12) {
|
||||
Text("Assign this line to a new speaker").font(.headline)
|
||||
TextField("Speaker name", text: $newSpeakerText).textFieldStyle(.roundedBorder).frame(width: 260)
|
||||
HStack {
|
||||
Spacer()
|
||||
Button("Cancel") { reassignNewIndex = nil }
|
||||
Button("Assign") {
|
||||
let n = newSpeakerText.trimmingCharacters(in: .whitespaces)
|
||||
if !n.isEmpty { model.reassign(index, to: n) }
|
||||
reassignNewIndex = nil
|
||||
}.keyboardShortcut(.defaultAction).disabled(newSpeakerText.trimmingCharacters(in: .whitespaces).isEmpty)
|
||||
}
|
||||
}
|
||||
.padding(16)
|
||||
}
|
||||
}
|
||||
|
||||
private struct IdentifiableString: Identifiable { let value: String; var id: String { value }; init(_ v: String) { value = v } }
|
||||
private struct IdentifiableInt: Identifiable { let value: Int; var id: Int { value }; init(_ v: Int) { value = v } }
|
||||
Reference in New Issue
Block a user