v0.9: VoicePipeline refactor — queue-based pipeline, lock/release model, thread-based TCP

Robovoice.App/MainForm.cs

@@ -17,10 +17,11 @@ internal sealed partial class MainForm : Form
     private LibPiperTtsEngine? _tts;
     private AudioOutput? _audioOutput;
     private TcpSttSource? _sttSource;
-    private Orchestrator? _orchestrator;
+    private VoicePipeline? _pipeline;
     private PttHotkey? _pttHotkey;
     private NotifyIcon? _trayIcon;
     private bool _trayInit;
+    private bool _locked;
 
     public MainForm()
     {
@@ -65,14 +66,11 @@ internal sealed partial class MainForm : Form
             chkMinimizeToTray.Checked = _config.MinimizeToTray;
             txtSttEndpoint.Text = _config.SttEndpoint;
 
-            btnBrowseFile.Click += OnBrowseFile;
-            btnTestVoice.Click += OnTestVoice;
             btnManageVoices.Click += OnManageVoices;
             btnClearLog.Click += (_, _) => txtLog.Clear();
+            btnLock.Click += OnLockToggle;
             txtPttKey.Enter += OnPttKeyFocus;
             txtPttKey.KeyDown += OnPttKeyDown;
-            cmbOutput.SelectedIndexChanged += OnOutputChanged;
-            cmbVoice.SelectedIndexChanged += OnVoiceChanged;
             txtSttEndpoint.Leave += OnSttEndpointChanged;
 
             trkNoise.Scroll += OnSliderScroll;
@@ -90,10 +88,8 @@ internal sealed partial class MainForm : Form
         }
         catch (Exception ex)
         {
-            System.Diagnostics.Debug.WriteLine($"OnLoad failed: {ex}");
+            Debug.WriteLine($"OnLoad failed: {ex}");
         }
-
-        BeginInvoke(async () => await InitializeEngineAsync());
     }
 
     private Keys _pttKey = Keys.F8;
@@ -216,7 +212,21 @@ internal sealed partial class MainForm : Form
             cmbOutput.SelectedIndex = 0;
     }
 
-    private async Task InitializeEngineAsync()
+    // ─── Lock / Release ────────────────────────────────────────────────────
+
+    private async void OnLockToggle(object? sender, EventArgs e)
+    {
+        if (_locked)
+        {
+            ReleaseModel();
+        }
+        else
+        {
+            await LockModelAsync();
+        }
+    }
+
+    private async Task LockModelAsync()
     {
         if (cmbVoice.SelectedItem is not string voiceName)
         {
@@ -237,44 +247,98 @@ internal sealed partial class MainForm : Form
             return;
         }
 
+        btnLock.Enabled = false;
         Log($"Loading voice: {voiceName}...");
-        _audioOutput?.Dispose();
-        if (_tts is not null)
-            await _tts.DisposeAsync();
-
-        _tts = new LibPiperTtsEngine(
-            modelPath,
-            _espeakDataPath,
-            noiseScale: trkNoise.Value / 1000.0f,
-            lengthScale: trkSpeed.Value / 100.0f,
-            noiseWScale: trkNoiseW.Value / 1000.0f);
-        _audioOutput = new AudioOutput();
 
         try
         {
-            if (_sttSource is null)
-            {
-                _sttSource = new TcpSttSource { Endpoint = txtSttEndpoint.Text, Log = Log };
-                await _sttSource.StartAsync();
-            }
+            _tts = new LibPiperTtsEngine(
+                modelPath,
+                _espeakDataPath,
+                noiseScale: trkNoise.Value / 1000.0f,
+                lengthScale: trkSpeed.Value / 100.0f,
+                noiseWScale: trkNoiseW.Value / 1000.0f);
 
-            if (_orchestrator is not null)
-                await _orchestrator.DisposeAsync();
-            _orchestrator = new Orchestrator(_tts, _audioOutput, _sttSource, Log)
+            await Task.Run(() => _tts.InitializeAsync());
+            Log($"TTS ready (sample rate: {_tts.SampleRate} Hz)");
+
+            _audioOutput = new AudioOutput();
+
+            _sttSource = new TcpSttSource { Endpoint = txtSttEndpoint.Text, Log = Log };
+            _sttSource.TranscriptReceived += OnTranscript;
+            await _sttSource.StartAsync();
+
+            _pipeline = new VoicePipeline(_tts, _audioOutput, Log)
             {
                 OutputDeviceName = cmbOutput.SelectedItem as string ?? string.Empty,
             };
+            _pipeline.Start();
+
+            _locked = true;
+            SetControlsLocked(true);
+            btnLock.Text = "Release Model";
+            btnLock.Enabled = true;
 
-            await _orchestrator.InitializeTtsAsync();
-            Log("Engine ready. Press PTT to send to STT server.");
             SetupHotkey();
+            Log("Model locked. Press PTT to talk.");
         }
         catch (Exception ex)
         {
-            Log($"Init failed: {ex.Message}");
+            Log($"Lock failed: {ex.Message}");
+            btnLock.Enabled = true;
+            _tts?.DisposeAsync().AsTask().Wait();
+            _tts = null;
+            _audioOutput?.Dispose();
+            _audioOutput = null;
+            if (_sttSource is not null)
+            {
+                _sttSource.TranscriptReceived -= OnTranscript;
+                _sttSource.DisposeAsync().AsTask().Wait();
+                _sttSource = null;
+            }
         }
     }
 
+    private void ReleaseModel()
+    {
+        if (!_locked) return;
+
+        _pttHotkey?.Dispose();
+        _pttHotkey = null;
+
+        if (_sttSource is not null)
+        {
+            _sttSource.TranscriptReceived -= OnTranscript;
+            _sttSource.DisposeAsync().AsTask().Wait();
+            _sttSource = null;
+        }
+
+        _pipeline?.Dispose();
+        _pipeline = null;
+
+        // VoicePipeline.Dispose calls _tts.DisposeAsync and _audioOutput.Dispose
+        _tts = null;
+        _audioOutput = null;
+
+        _locked = false;
+        SetControlsLocked(false);
+        btnLock.Text = "Lock Model";
+        Log("Model released.");
+    }
+
+    private void SetControlsLocked(bool locked)
+    {
+        cmbVoice.Enabled = !locked;
+        cmbOutput.Enabled = !locked;
+        trkNoise.Enabled = !locked;
+        trkSpeed.Enabled = !locked;
+        trkNoiseW.Enabled = !locked;
+        txtSttEndpoint.Enabled = !locked;
+        btnManageVoices.Enabled = !locked;
+    }
+
+    // ─── PTT ────────────────────────────────────────────────────────────────
+
     private void SetupHotkey()
     {
         _pttHotkey?.Dispose();
@@ -287,86 +351,56 @@ internal sealed partial class MainForm : Form
 
     private void OnPttPressed(object? sender, EventArgs e)
     {
-        Log("PTT pressed");
-        _orchestrator?.NotifyPttPressed();
-        try
-        {
-            _sttSource?.SendOn();
-        }
-        catch (Exception ex)
-        {
-            Log($"SendOn failed: {ex.Message}");
-        }
+        _pipeline?.OnPttPressed();
+        _sttSource?.SendOn();
     }
 
     private void OnPttReleased(object? sender, EventArgs e)
     {
-        Log("PTT released");
-        try
-        {
-            _sttSource?.SendOff();
-        }
-        catch (Exception ex)
-        {
-            Log($"SendOff failed: {ex.Message}");
-        }
+        _sttSource?.SendOff();
+        _pipeline?.OnPttReleased();
     }
 
-    private void OnBrowseFile(object? sender, EventArgs e)
+    // ─── STT transcript → pipeline ──────────────────────────────────────────
+
+    private void OnTranscript(object? sender, TranscriptEventArgs e)
     {
-        using var dlg = new OpenFileDialog
-        {
-            Filter = "Text files (*.txt)|*.txt|All files (*.*)|*.*",
-            Title = "Select a text file to speak",
-        };
+        if (_pipeline is null || !_locked) return;
 
-        if (dlg.ShowDialog() == DialogResult.OK)
+        var msg = e.Message;
+        if (msg.Type == TranscriptType.Partial)
         {
-            string text = File.ReadAllText(dlg.FileName);
-            lblFileName.Text = Path.GetFileName(dlg.FileName);
-            lblFileName.ForeColor = Color.Black;
-            Log($"Loaded: {dlg.FileName} ({text.Length} chars)");
-            _ = _orchestrator?.SynthesizeAsync(text);
+            if (!string.IsNullOrWhiteSpace(msg.Text))
+            {
+                Log($"SEGMENT: \"{msg.Text}\"");
+                _pipeline.EnqueueSegment(msg.Text);
+            }
+        }
+        else
+        {
+            if (!string.IsNullOrWhiteSpace(msg.Text))
+            {
+                Log($"FINAL: \"{msg.Text}\"");
+                _pipeline.EnqueueFinal(msg.Text);
+            }
+            else
+            {
+                Log("FINAL: (empty)");
+                _pipeline.EnqueueFinal("");
+            }
         }
     }
 
-    private void OnTestVoice(object? sender, EventArgs e)
-    {
-        if (_orchestrator is null)
-        {
-            Log("Engine not initialized.");
-            return;
-        }
-
-        if (cmbVoice.SelectedItem is not string voiceName)
-        {
-            Log("No voice selected.");
-            return;
-        }
-
-        if (!VoiceCatalogue.IsVoiceInstalled(_voicesDir, voiceName))
-        {
-            Log($"Voice '{voiceName}' is not installed. Use Add/Remove to download it.");
-            return;
-        }
-
-        string text = txtTextInput.Text.Trim();
-        if (string.IsNullOrEmpty(text))
-            text = "Hello, this is a voice test.";
-
-        btnTestVoice.Enabled = false;
-        try
-        {
-            _ = _orchestrator.SynthesizeAsync(text);
-        }
-        finally
-        {
-            btnTestVoice.Enabled = true;
-        }
-    }
+    // ─── Voice manager ─────────────────────────────────────────────────────
 
     private void OnManageVoices(object? sender, EventArgs e)
     {
+        if (_locked)
+        {
+            Log("Release the model before managing voices.");
+            return;
+        }
+
         string currentVoice = cmbVoice.SelectedItem as string ?? string.Empty;
         using var dlg = new VoiceManagerForm(_voicesDir, currentVoice);
         dlg.ShowDialog(this);
@@ -384,31 +418,9 @@ internal sealed partial class MainForm : Form
         }
 
         SaveConfig();
-        _ = InitializeEngineAsync();
     }
 
-    private void OnVoiceChanged(object? sender, EventArgs e)
-    {
-        btnTestVoice.Enabled = cmbVoice.SelectedItem is string voiceName
-            && VoiceCatalogue.IsVoiceInstalled(_voicesDir, voiceName);
-
-        if (cmbVoice.SelectedItem is string name)
-        {
-            SaveConfig();
-            _ = ReinitializeEngineAsync(name);
-        }
-    }
-
-    private async Task ReinitializeEngineAsync(string voiceName)
-    {
-        if (!VoiceCatalogue.IsVoiceInstalled(_voicesDir, voiceName))
-        {
-            btnTestVoice.Enabled = false;
-            return;
-        }
-
-        await InitializeEngineAsync();
-    }
+    // ─── Slider / endpoint ─────────────────────────────────────────────────
 
     private void OnSliderScroll(object? sender, EventArgs e)
     {
@@ -420,33 +432,16 @@ internal sealed partial class MainForm : Form
     private void OnSliderReleased(object? sender, MouseEventArgs e)
     {
         SaveConfig();
-        if (cmbVoice.SelectedItem is string name && VoiceCatalogue.IsVoiceInstalled(_voicesDir, name))
-        {
-            _ = InitializeEngineAsync();
-        }
     }
 
-    private void OnOutputChanged(object? sender, EventArgs e)
-    {
-        if (_orchestrator is not null)
-            _orchestrator.OutputDeviceName = cmbOutput.SelectedItem as string ?? string.Empty;
-        Log($"Output device: {_orchestrator?.OutputDeviceName}");
-        SaveConfig();
-    }
-
-    private async void OnSttEndpointChanged(object? sender, EventArgs e)
+    private void OnSttEndpointChanged(object? sender, EventArgs e)
     {
         _config.SttEndpoint = txtSttEndpoint.Text;
         SaveConfig();
-
-        if (_sttSource is not null)
-        {
-            await _sttSource.DisposeAsync();
-            _sttSource = null;
-            await InitializeEngineAsync();
-        }
     }
 
+    // ─── Tray ──────────────────────────────────────────────────────────────
+
     private void SetupTray()
     {
         if (_trayInit) return;
@@ -503,6 +498,8 @@ internal sealed partial class MainForm : Form
         Activate();
     }
 
+    // ─── Config ────────────────────────────────────────────────────────────
+
     private void SaveConfig()
     {
         _config.Voice = cmbVoice.SelectedItem as string ?? string.Empty;
@@ -516,6 +513,8 @@ internal sealed partial class MainForm : Form
         _config.Save();
     }
 
+    // ─── Logging ───────────────────────────────────────────────────────────
+
     private void Log(string message)
     {
         if (IsDisposed) return;
@@ -530,12 +529,35 @@ internal sealed partial class MainForm : Form
         txtLog.ScrollToCaret();
     }
 
+    // ─── Shutdown ──────────────────────────────────────────────────────────
+
+    private static string DisposeLogPath => Path.Combine(AppConfig.AppDataDir, "dispose.log");
+
+    private static void LogDispose(string label, long elapsedMs)
+    {
+        string line = $"[{DateTime.Now:HH:mm:ss.fff}] {label}: {elapsedMs}ms";
+        try
+        {
+            Directory.CreateDirectory(AppConfig.AppDataDir);
+            File.AppendAllText(DisposeLogPath, line + "\n");
+        }
+        catch { }
+    }
+
     private void OnFormClosing(object? sender, FormClosingEventArgs e)
     {
+        var sw = Stopwatch.StartNew();
         _pttHotkey?.Dispose();
+        LogDispose("pttHotkey.Dispose", sw.ElapsedMilliseconds);
+
         _trayIcon!.Visible = false;
-        _orchestrator?.DisposeAsync().AsTask().Wait(2000);
-        _sttSource?.DisposeAsync().AsTask().Wait(2000);
+
+        sw.Restart();
+        ReleaseModel();
+        LogDispose("ReleaseModel", sw.ElapsedMilliseconds);
+
+        sw.Restart();
         SaveConfig();
+        LogDispose("SaveConfig", sw.ElapsedMilliseconds);
     }
 }