Files
voice-cat/clients/apple/iOS/VoiceCatiOS/BroadcastAudioPump.swift
Talon 8c90e250f0 feat(apple): screen-audio sharing -- macOS ScreenCaptureKit, iOS ReplayKit
Implement system/desktop audio sharing on the Apple clients, feeding the
existing SCREEN_AUDIO Opus -> AEAD -> UDP path via vc_stream_feed_pcm. No
C++/protocol/codec changes -- the core was already ready (the Windows-only
loopback is #ifdef VOICECAT_HAS_LOOPBACK; off Windows the stream just waits
for fed PCM). Audio only; video is dropped.

macOS (in-process):
- ScreenAudioCapture.swift drives an audio-only SCStream
  (excludesCurrentProcessAudio), converts Float32 -> int16 in the channel's
  mono/stereo mode, and calls feedPcm. Capture starts on the self
  .streamStarted event (effective config known then). Wired into
  MainWindowController.screenAudioClicked().

iOS (forward-to-host, single session):
- VoiceCatBroadcast: a ReplayKit Broadcast Upload Extension consumes
  .audioApp only, resamples to 48kHz int16 stereo (AVAudioConverter), and
  writes a shared App Group SPSC ring (BroadcastAudioRing.swift). It does
  not link libvoicecat.
- Host BroadcastAudioPump drains the ring (reacting to the extension's
  Darwin notifications) and feeds the SCREEN_AUDIO stream it owns, downmixing
  to mono when the channel is mono. Screen audio appears as a second stream
  of the same user; no credentials persisted. UI is RPSystemBroadcastPicker
  View in VoiceControlsView. Removes the speculative BroadcastCredentials.

Docs: voice.md s9, CLAUDE.md status, PROGRESS.md.
2026-06-21 00:14:31 +02:00

160 lines
6.5 KiB
Swift

import Foundation
// BroadcastAudioPump host side of iOS screen-audio sharing (docs/voice.md §9, iOS detail).
//
// Drains the App Group shared-memory ring written by the broadcast upload extension and hands
// whole 20 ms frames to a feed closure (which calls VoiceCatClient.feedPcm). The host app owns
// the SCREEN_AUDIO stream, so screen audio appears as a second stream of the SAME user exactly
// like the Windows/macOS desktop-audio share, and a single session (no separate connection).
//
// It reacts to the extension's Darwin notifications for prompt start/stop, and the drain timer
// also watches the ring's active flag as a safety net if a notification is missed. The ring
// always carries stereo; we downmix to mono when the stream's effective config is mono.
//
// Not @MainActor: the drain runs on a background queue (feedPcm is thread-safe). The start/stop
// callbacks are dispatched to main so they can drive @MainActor SessionState.
final class BroadcastAudioPump {
/// The host should start the SCREEN_AUDIO stream (a broadcast became active).
var onBroadcastStarted: (() -> Void)?
/// The host should stop the SCREEN_AUDIO stream (the broadcast ended).
var onBroadcastFinished: (() -> Void)?
private static let frameSamplesPerChannel = 960 // 20 ms @ 48 kHz
private let queue = DispatchQueue(label: "cat.voice.broadcast.pump")
private var ring: BroadcastAudioRing?
private var timer: DispatchSourceTimer?
private var feed: ((UnsafePointer<Int16>, Int, UInt32) -> Void)?
private var streamChannels = 1
private var ringChannels = 2
private var pending: [Int16] = [] // interleaved at ringChannels width
private var scratch = [Int16](repeating: 0, count: 8192)
private var started = false
// MARK: - Lifecycle (host connect/disconnect)
func start() {
guard !started else { return }
ring = try? BroadcastAudioRing()
started = true
registerDarwin()
// Host reconnected while a broadcast is still running pick it up.
if ring?.isActive == true { onBroadcastStarted?() }
}
func stop() {
guard started else { return }
started = false
unregisterDarwin()
endFeeding()
ring = nil
}
// MARK: - Feeding (driven by the host once the stream is live)
/// Begin draining the ring into `feed`. Called after the SCREEN_AUDIO stream's
/// `.streamStarted` event, when its effective channel count is known.
func beginFeeding(streamChannels: UInt32,
feed: @escaping (UnsafePointer<Int16>, Int, UInt32) -> Void) {
queue.async {
self.streamChannels = max(1, min(2, Int(streamChannels)))
self.ringChannels = max(1, Int(self.ring?.channels ?? 2))
self.feed = feed
self.pending.removeAll(keepingCapacity: true)
self.ring?.drainStale() // discard pre-roll buffered before we were ready
self.startTimer()
}
}
func endFeeding() {
queue.async {
self.timer?.cancel()
self.timer = nil
self.feed = nil
self.pending.removeAll(keepingCapacity: true)
}
}
// MARK: - Drain
private func startTimer() {
let t = DispatchSource.makeTimerSource(queue: queue)
t.schedule(deadline: .now(), repeating: .milliseconds(10), leeway: .milliseconds(2))
t.setEventHandler { [weak self] in self?.drain() }
timer = t
t.resume()
}
private func drain() {
guard let ring, let feed else { return }
// Safety net: broadcast ended but we missed the Darwin note.
if !ring.isActive {
DispatchQueue.main.async { [weak self] in self?.onBroadcastFinished?() }
return
}
while true {
let got = scratch.withUnsafeMutableBufferPointer { ring.read(into: $0) }
if got == 0 { break }
pending.append(contentsOf: scratch[0..<got])
if got < scratch.count { break }
}
emitFrames(feed)
}
private func emitFrames(_ feed: (UnsafePointer<Int16>, Int, UInt32) -> Void) {
let n = Self.frameSamplesPerChannel
let rc = ringChannels
let sc = streamChannels
let inFrame = n * rc
while pending.count >= inFrame {
if sc == rc {
pending.withUnsafeBufferPointer { feed($0.baseAddress!, n, UInt32(sc)) }
} else if sc == 1 && rc == 2 {
var mono = [Int16](repeating: 0, count: n)
pending.withUnsafeBufferPointer { buf in
let p = buf.baseAddress!
for i in 0..<n { mono[i] = Int16((Int(p[i * 2]) + Int(p[i * 2 + 1])) / 2) }
}
mono.withUnsafeBufferPointer { feed($0.baseAddress!, n, 1) }
} else if sc == 2 && rc == 1 {
var stereo = [Int16](repeating: 0, count: n * 2)
pending.withUnsafeBufferPointer { buf in
let p = buf.baseAddress!
for i in 0..<n { stereo[i * 2] = p[i]; stereo[i * 2 + 1] = p[i] }
}
stereo.withUnsafeBufferPointer { feed($0.baseAddress!, n, 2) }
}
pending.removeFirst(inFrame)
}
}
// MARK: - Darwin notifications
private func registerDarwin() {
let observer = Unmanaged.passUnretained(self).toOpaque()
let center = CFNotificationCenterGetDarwinNotifyCenter()
let callback: CFNotificationCallback = { _, observer, name, _, _ in
guard let observer, let name else { return }
let pump = Unmanaged<BroadcastAudioPump>.fromOpaque(observer).takeUnretainedValue()
let raw = name.rawValue as String
DispatchQueue.main.async {
if raw == BroadcastNotification.started { pump.onBroadcastStarted?() }
else if raw == BroadcastNotification.finished { pump.onBroadcastFinished?() }
}
}
CFNotificationCenterAddObserver(center, observer, callback,
BroadcastNotification.started as CFString, nil, .deliverImmediately)
CFNotificationCenterAddObserver(center, observer, callback,
BroadcastNotification.finished as CFString, nil, .deliverImmediately)
}
private func unregisterDarwin() {
CFNotificationCenterRemoveEveryObserver(
CFNotificationCenterGetDarwinNotifyCenter(),
Unmanaged.passUnretained(self).toOpaque())
}
deinit { if started { unregisterDarwin() } }
}