feat(flow): add whole-utterance batch ASR fallback (P1)

- Accumulate utterance PCM in FlowContinuousCapture for batch retry - Run full-utterance transcribeChunk when stitched final lags partial - Refactor Mac MLX tail drain to shared FlowUtteranceEndCoordinator - Add FlowUtterancePCMStore, UtteranceBatchFallbackPolicy, and tests Co-authored-by: Rocky <hkgood@users.noreply.github.com>
2026-07-26 10:15:31 +00:00
parent c7ee891a90
commit f0240c3088
9 changed files with 256 additions and 3 deletions
@@ -0,0 +1,41 @@
+// UtteranceBatchFallbackPolicy.swift
+// OSGKeyboard · Shared
+//
+// Decides when to re-run ASR on the full utterance PCM after pipelined
+// chunking, and how to pick the best transcript among candidates.
+
+import Foundation
+
+public enum UtteranceBatchFallbackPolicy {
+    public static let defaultCharacterAdvantage = UtteranceTranscriptGuard.defaultPartialAdvantage
+
+    /// True when chunked output likely lost trailing content vs the live partial.
+    public static func shouldRunBatchFallback(
+        stitchedFinal: String,
+        partialSnapshot: String,
+        minimumCharacterAdvantage: Int = defaultCharacterAdvantage
+    ) -> Bool {
+        let final = stitchedFinal.trimmingCharacters(in: .whitespacesAndNewlines)
+        let partial = partialSnapshot.trimmingCharacters(in: .whitespacesAndNewlines)
+
+        if final.isEmpty, !partial.isEmpty { return true }
+        if partial.isEmpty { return false }
+        return partial.count >= final.count + minimumCharacterAdvantage
+    }
+
+    /// Prefer the longest non-empty transcript after batch ASR completes.
+    public static func preferredTranscript(
+        batch: String,
+        stitchedFinal: String,
+        partialSnapshot: String,
+        current: String
+    ) -> String {
+        let candidates = [batch, current, stitchedFinal, partialSnapshot]
+            .map { $0.trimmingCharacters(in: .whitespacesAndNewlines) }
+            .filter { !$0.isEmpty }
+        guard let best = candidates.max(by: { $0.count < $1.count }) else {
+            return current.trimmingCharacters(in: .whitespacesAndNewlines)
+        }
+        return best
+    }
+}