Phase 1: dual-track audio capture → mixed-mono 16 kHz WAV + mic VAD

AudioRecorder captures system audio (ScreenCaptureKit) + mic (AVAudioEngine) on a single serial ioQueue, one shared monotonic t0, time-driven writers (pad gaps / trim overlaps) so tracks stay aligned, and an energy mic-VAD for 'self' spans. AudioMixer sums the aligned tracks into mixed_mono_16k.wav. SessionController drives a serialized start/stop state machine, writes the session folder + self_vad.json, exposes live level meters, and finalizes on quit. Hardening from review: ioQueue single-domain (no races), stop() never hangs (mic-first teardown + bounded stopCapture), layout-agnostic mic deep-copy, discard-only video output to keep SCStream alive, VAD lockstep on committed frames, stable signing team in project.yml, single-instance enforcement.
2026-06-05 21:30:11 -05:00
parent b2ae3a62b9
commit fd7e1a5907
12 changed files with 1018 additions and 10 deletions
@@ -0,0 +1,67 @@
+import AVFoundation
+
+/// Sums the two aligned 16 kHz mono tracks (mic + system) into the single
+/// **mixed-mono 16 kHz WAV** that the backend receives. Both inputs are already
+/// front-padded to the shared t0, so frame N of each file is the same instant.
+/// Streamed in 1-second chunks to keep memory flat for long calls.
+enum AudioMixer {
+    static func mix(mic micURL: URL, system systemURL: URL, into outURL: URL) throws {
+        let mic = try? AVAudioFile(forReading: micURL)
+        let sys = try? AVAudioFile(forReading: systemURL)
+
+        let settings: [String: Any] = [
+            AVFormatIDKey: kAudioFormatLinearPCM,
+            AVSampleRateKey: 16_000,
+            AVNumberOfChannelsKey: 1,
+            AVLinearPCMBitDepthKey: 16,
+            AVLinearPCMIsFloatKey: false,
+            AVLinearPCMIsBigEndianKey: false,
+        ]
+        let out = try AVAudioFile(
+            forWriting: outURL,
+            settings: settings,
+            commonFormat: .pcmFormatFloat32,
+            interleaved: false)
+
+        let outFormat = Resampler.targetFormat
+        let chunk: AVAudioFramePosition = 16_000
+        let total = max(mic?.length ?? 0, sys?.length ?? 0)
+        var pos: AVAudioFramePosition = 0
+
+        while pos < total {
+            let frames = AVAudioFrameCount(min(chunk, total - pos))
+            guard let mixBuf = AVAudioPCMBuffer(pcmFormat: outFormat, frameCapacity: frames),
+                  let dst = mixBuf.floatChannelData?[0] else { break }
+            mixBuf.frameLength = frames
+            memset(dst, 0, Int(frames) * MemoryLayout<Float>.size)
+
+            add(file: mic, at: pos, maxFrames: frames, into: dst)
+            add(file: sys, at: pos, maxFrames: frames, into: dst)
+
+            var i = 0
+            while i < Int(frames) {
+                if dst[i] > 1 { dst[i] = 1 } else if dst[i] < -1 { dst[i] = -1 }
+                i += 1
+            }
+            try out.write(from: mixBuf)
+            pos += AVAudioFramePosition(frames)
+        }
+    }
+
+    private static func add(file: AVAudioFile?, at pos: AVAudioFramePosition,
+                            maxFrames: AVAudioFrameCount, into dst: UnsafeMutablePointer<Float>) {
+        guard let file, pos < file.length else { return }
+        file.framePosition = pos
+        let toRead = AVAudioFrameCount(min(AVAudioFramePosition(maxFrames), file.length - pos))
+        guard toRead > 0,
+              let buf = AVAudioPCMBuffer(pcmFormat: file.processingFormat, frameCapacity: toRead)
+        else { return }
+        do {
+            try file.read(into: buf, frameCount: toRead)
+            guard let src = buf.floatChannelData?[0] else { return }
+            var i = 0
+            let count = Int(buf.frameLength)
+            while i < count { dst[i] += src[i]; i += 1 }
+        } catch {}
+    }
+}