Grant Gilliam
863136aeec
Phase 2 (call detection): CallDetector using CoreAudio per-process mic attribution (anarlog technique) — robust start+stop for Zoom/Teams/Signal/Meet, ignoring our own recording; auto-record toggle. Built; pending live multi-app confirmation by the user. Phase 3 (visual timeline foundation): AppAdapter protocol + SpeakerObservation, TimelineBuilder (hysteresis/overlap/self-merge/aliases), VisualTimeline (schema 1.1), TextRecognizer (Vision OCR), FrameSampler + GridCallAnalyzer (name OCR + saturated-highlight active-speaker attribution), SignalAdapter, VisualObserver (window capture; frames released, never saved; minimized->visual_gap, idle != gap). Synthetic-frame tested; adapter geometry pending real Signal fixtures + live VisualObserver validation. Phase 5 (backend hand-off): SparkControlClient (multipart label-merge, sequential, TLS-skip, 503 Retry-After/413), SessionPackager (chunk plan + WAV slice + timeline slice/rebase), TranscriptAssembler + SpeakersFile, TranscriptPipeline. Validated END-TO-END against the live backend (chunk -> label-merge -> speakers.json). Phase 6 (voiceprints): VoiceprintStore (known_voiceprints, persist named fingerprints, skip Unknown). Wired: 'Send to backend' button + transcript status, auto-send toggle (default off) + self-name setting. All adversarial-review findings fixed. App + XCTest suite build; tests pass.
61 lines
2.2 KiB
Swift
61 lines
2.2 KiB
Swift
import XCTest
|
|
@testable import Ten31Transcripts
|
|
|
|
final class TimelineBuilderTests: XCTestCase {
|
|
private func obs(_ name: String, _ speaking: Bool, _ t: Double, _ conf: Double = 0.9) -> SpeakerObservation {
|
|
SpeakerObservation(name: name, speaking: speaking, bbox: .zero, confidence: conf, t: t)
|
|
}
|
|
|
|
func testOpensAfterKFramesAndClosesAfterMQuiet() {
|
|
let b = TimelineBuilder(openFrames: 2, closeFrames: 2)
|
|
b.ingest([obs("A", true, 0)], at: 0)
|
|
b.ingest([obs("A", true, 1)], at: 1)
|
|
b.ingest([obs("A", true, 2)], at: 2)
|
|
b.ingest([], at: 3)
|
|
b.ingest([], at: 4)
|
|
b.finish()
|
|
XCTAssertEqual(b.segments.count, 1)
|
|
XCTAssertEqual(b.segments.first?.name, "A")
|
|
XCTAssertEqual(b.segments.first?.start ?? -1, 0, accuracy: 0.001)
|
|
XCTAssertEqual(b.segments.first?.end ?? -1, 2, accuracy: 0.001)
|
|
XCTAssertEqual(b.segments.first?.source, "vision")
|
|
}
|
|
|
|
func testSingleFlickerDoesNotOpen() {
|
|
let b = TimelineBuilder(openFrames: 2, closeFrames: 2)
|
|
b.ingest([obs("A", true, 0)], at: 0)
|
|
b.ingest([], at: 1)
|
|
b.finish()
|
|
XCTAssertTrue(b.segments.isEmpty)
|
|
}
|
|
|
|
func testAllowsOverlap() {
|
|
let b = TimelineBuilder(openFrames: 1, closeFrames: 1)
|
|
b.ingest([obs("A", true, 0), obs("B", true, 0)], at: 0)
|
|
b.ingest([obs("A", true, 1), obs("B", true, 1)], at: 1)
|
|
b.ingest([], at: 2)
|
|
b.finish()
|
|
XCTAssertEqual(b.segments.count, 2)
|
|
XCTAssertEqual(Set(b.segments.map { $0.name }), ["A", "B"])
|
|
}
|
|
|
|
func testMergesSelfSpans() {
|
|
let b = TimelineBuilder()
|
|
b.mergeSelfSpans([VADSpan(start: 0, end: 4.5, confidence: 0.97)], selfName: "Grant")
|
|
b.finish()
|
|
XCTAssertEqual(b.segments.count, 1)
|
|
XCTAssertEqual(b.segments.first?.name, "Grant")
|
|
XCTAssertEqual(b.segments.first?.source, "mic_vad")
|
|
}
|
|
|
|
func testNormalizesAlias() {
|
|
let b = TimelineBuilder(openFrames: 1, closeFrames: 1)
|
|
b.addAlias("Sarah J", canonical: "Sarah Jones")
|
|
b.ingest([obs("Sarah J", true, 0)], at: 0)
|
|
b.ingest([obs("Sarah J", true, 1)], at: 1)
|
|
b.ingest([], at: 2)
|
|
b.finish()
|
|
XCTAssertEqual(b.segments.first?.name, "Sarah Jones")
|
|
}
|
|
}
|