Added Model Selection for Elevenlabs _ Added Voice Styles for Elevenlabs, Azure, and OpenAI. Minor UI bugfixes

src/TextToTalk.Lexicons/LexiconManager.cs

@@ -2,6 +2,7 @@
 using System.Collections.Generic;
 using System.IO;
 using System.Linq;
+using System.Text.RegularExpressions;
 using System.Xml.Linq;
 
 namespace TextToTalk.Lexicons;
@@ -109,6 +110,26 @@ public string MakeSsml(
                 }
             }
         }
+        bool hasStyles = text.Contains("[") && text.Contains("]");
+        if (hasStyles)
+        {
+            // This regex captures the style name in group 1 and the text in group 2.
+            // It replaces the whole match with the SSML tag, effectively removing 
+            // the [styleName] text from the spoken output.
+            text = Regex.Replace(text, @"\[\[(.*?)\]\]", m =>
+            {
+                var styleName = m.Groups[1].Value.Trim();
+                var content = m.Groups[2].Value; // Captured text after the bracket
+                if (voice != null)
+                {
+                    return $"<mstts:express-as style=\"{styleName}\" styledegree=\"1.5\">{content}</mstts:express-as>";
+                }
+                else
+                {
+                    return content; // If no voice is specified, just return the content without styling.  This ensures System.Speech compatibility.
+                }
+            });
+        }
 
         if (playbackRate >= 0)
         {
@@ -125,7 +146,10 @@ public string MakeSsml(
         var speakTag = "<speak";
         if (includeSpeakAttributes)
         {
-            speakTag += " version=\"1.0\" xmlns=\"http://www.w3.org/2001/10/synthesis\"";
+            // Correctly define both the default SSML namespace and the Microsoft-specific namespace
+            speakTag += " version=\"1.0\" " +
+                        "xmlns=\"http://www.w3.org/2001/10/synthesis\" " +
+                        "xmlns:mstts=\"http://www.w3.org/2001/mstts\""; // Fixed URI
 
             if (langCode != null)
             {

src/TextToTalk/Backends/Azure/AzureBackend.cs

@@ -1,13 +1,17 @@
-using System;
+using Dalamud.Bindings.ImGui;
+using FFXIVClientStructs.FFXIV.Client.Game.UI;
+using System;
+using System.Collections.Generic;
 using System.Net.Http;
-using Dalamud.Bindings.ImGui;
+using static TextToTalk.Backends.Azure.AzureClient;
 
 namespace TextToTalk.Backends.Azure;
 
 public class AzureBackend : VoiceBackend
 {
     private readonly AzureBackendUI ui;
     private readonly AzureBackendUIModel uiModel;
+    public List<VoiceDetails> voices;
 
     public AzureBackend(PluginConfiguration config, HttpClient http)
     {
@@ -17,7 +21,14 @@ public AzureBackend(PluginConfiguration config, HttpClient http)
         LexiconUtils.LoadFromConfigAzure(lexiconManager, config);
 
         this.uiModel = new AzureBackendUIModel(config, lexiconManager);
+        this.voices = this.uiModel.voices;
         this.ui = new AzureBackendUI(this.uiModel, config, lexiconManager, http, this);
+
+    }
+
+    public override void DrawStyles(IConfigUIDelegates helpers)
+    {
+        helpers.OpenVoiceStylesConfig();
     }
 
     public override void Say(SayRequest request)
@@ -82,4 +93,5 @@ protected override void Dispose(bool disposing)
             this.uiModel.Azure?.Dispose();
         }
     }
+
 }

src/TextToTalk/Backends/Azure/AzureBackendUI.cs

@@ -1,7 +1,6 @@
 using Dalamud.Bindings.ImGui;
 using Dalamud.Game;
 using Dalamud.Game.Text;
-using Google.Api;
 using System;
 using System.IO;
 using System.Linq;
@@ -10,6 +9,7 @@
 using TextToTalk.Lexicons.Updater;
 using TextToTalk.UI;
 using TextToTalk.UI.Lexicons;
+using TextToTalk.UI.Windows;
 
 namespace TextToTalk.Backends.Azure;
 
@@ -87,6 +87,10 @@ public void DrawSettings(IConfigUIDelegates helpers)
         {
             ImGui.TextColored(ImColor.Red, "You have no presets. Please create one using the \"New preset\" button.");
         }
+        else if (currentVoicePreset == null && presets.Count > 0)
+        {
+            config.SetCurrentVoicePreset(presets.First().Id);
+        }
 
         BackendUI.NewPresetButton<AzureVoicePreset>($"New preset##{MemoizedId.Create()}", this.config);
 
@@ -111,7 +115,7 @@ public void DrawSettings(IConfigUIDelegates helpers)
 
         {
             var voices = this.model.Voices;
-            string?[] voiceArray = voices.ToArray();
+            string?[] voiceArray = voices.Where(v => v != null && !string.IsNullOrEmpty(v.Name)).Select(v => v.Name).ToArray();
             var voiceIndex = Array.IndexOf(voiceArray, currentVoicePreset.VoiceName);
             if (ImGui.Combo($"Voice##{MemoizedId.Create()}", ref voiceIndex, voiceArray, voices.Count))
             {
@@ -126,7 +130,7 @@ public void DrawSettings(IConfigUIDelegates helpers)
                         "No voices are available on this voice engine for the current region.\n" +
                         "Please log in using a different region.");
                     break;
-                case > 0 when !voices.Any(v => v == currentVoicePreset.VoiceName):
+                case > 0 when !voiceArray.Any(v => v == currentVoicePreset.VoiceName):
                     BackendUI.ImGuiVoiceNotSelected();
                     break;
             }
@@ -167,6 +171,11 @@ public void DrawSettings(IConfigUIDelegates helpers)
                 backend.Say(request);
             }
         }
+        ImGui.SameLine();
+        if (ImGui.Button($"Voice Styles##{MemoizedId.Create()}"))
+        {
+            VoiceStyles.Instance?.ToggleStyle();
+        }
 
         this.lexiconComponent.Draw();
         ImGui.Spacing();

src/TextToTalk/Backends/Azure/AzureBackendUIModel.cs

@@ -2,6 +2,7 @@
 using System.Collections.Generic;
 using System.Text.RegularExpressions;
 using TextToTalk.Lexicons;
+using static TextToTalk.Backends.Azure.AzureClient;
 
 namespace TextToTalk.Backends.Azure;
 
@@ -12,7 +13,7 @@ public class AzureBackendUIModel
     private readonly PluginConfiguration config;
     private readonly LexiconManager lexiconManager;
 
-    private List<string> voices;
+    public List<VoiceDetails> voices;
     private AzureLoginInfo loginInfo;
 
     /// <summary>
@@ -28,13 +29,13 @@ public class AzureBackendUIModel
     /// <summary>
     /// Gets the available voices.
     /// </summary>
-    public IReadOnlyList<string> Voices => this.voices;
+    public IReadOnlyList<VoiceDetails> Voices => this.voices;
 
     public AzureBackendUIModel(PluginConfiguration config, LexiconManager lexiconManager)
     {
         this.config = config;
         this.lexiconManager = lexiconManager;
-        this.voices = new List<string>();
+        this.voices = new List<VoiceDetails>();
 
         this.loginInfo = new AzureLoginInfo();
         var credentials = AzureCredentialManager.LoadCredentials();
@@ -98,7 +99,7 @@ private bool TryAzureLogin()
             DetailedLog.Info($"Logging into Azure region {this.loginInfo.Region}");
             Azure = new AzureClient(this.loginInfo.SubscriptionKey, this.loginInfo.Region, this.lexiconManager, this.config);
             // This should throw an exception if the login failed
-            this.voices = Azure.GetVoices();
+            this.voices = Azure.GetVoicesWithStyles();
             return true;
         }
         catch (Exception e)

src/TextToTalk/Backends/Azure/AzureClient.cs

@@ -30,6 +30,25 @@ public TextSource GetCurrentlySpokenTextSource()
     {
         return this.soundQueue.GetCurrentlySpokenTextSource();
     }
+    public List<VoiceDetails> GetVoicesWithStyles()
+    {
+        // Fetches the voice result asynchronously and waits for completion
+        var res = this.synthesizer.GetVoicesAsync().GetAwaiter().GetResult();
+        HandleResult(res);
+
+        // Maps each voice to a custom object containing Name and StyleList
+        return res.Voices.Select(voice => new VoiceDetails
+        {
+            Name = voice.Name,
+            Styles = voice.StyleList.ToList() // StyleList is a string[]
+        }).ToList();
+    }
+
+    public class VoiceDetails
+    {
+        public string Name { get; set; }
+        public List<string> Styles { get; set; }
+    }
 
     public List<string> GetVoices()
     {

src/TextToTalk/Backends/Azure/AzureVoiceStylesUI.cs

@@ -0,0 +1,63 @@
+using Dalamud.Bindings.ImGui;
+using System.Linq;
+using TextToTalk;
+using TextToTalk.Backends;
+using TextToTalk.Backends.Azure;
+using TextToTalk.UI.Windows;
+using static TextToTalk.Backends.Azure.AzureClient;
+
+public class AzureVoiceStyles : IVoiceStylesWindow
+{
+    private readonly AzureBackend backend;
+    private PluginConfiguration config;
+    private VoiceStyles voiceStyles;
+    static double lastCopyTime = -1.0;
+    static string lastCopiedStyle = "";
+
+    public AzureVoiceStyles(AzureBackend backend, PluginConfiguration config, VoiceStyles voiceStyles)
+    {
+        this.backend = backend;
+        this.config = config;
+        this.voiceStyles = voiceStyles;
+
+    }
+
+    public void Draw(IConfigUIDelegates helpers)
+    {
+        var currentVoicePreset = this.config.GetCurrentVoicePreset<AzureVoicePreset>();
+        var voiceDetails = this.backend.voices
+            .OrderBy(v => v.Name)
+            .FirstOrDefault(v => v?.Name == currentVoicePreset?.VoiceName);
+
+        if (voiceDetails?.Styles == null || voiceDetails.Styles.Count == 0)
+        {
+            ImGui.TextDisabled("No styles available for this voice.");
+            return;
+        }
+
+        ImGui.Text("Click a style to copy its tag to clipboard:");
+        ImGui.Separator();
+
+        foreach (var style in voiceDetails.Styles)
+        {
+            if (string.IsNullOrEmpty(style)) continue;
+
+            if (ImGui.Selectable(style))
+            {
+                VoiceStyles.Instance?.CopyStyleToClipboard(style);
+                lastCopyTime = ImGui.GetTime();
+                lastCopiedStyle = style;
+            }
+
+            if (lastCopiedStyle == style && (ImGui.GetTime() - lastCopyTime < 1.0))
+            {
+                ImGui.SetTooltip("Copied!");
+            }
+            else if (ImGui.IsItemHovered())
+            {
+                ImGui.SetTooltip($"Click to copy");
+            }
+
+        }
+    }
+}

src/TextToTalk/Backends/BackendUI.cs

@@ -73,9 +73,21 @@ public static void DeletePresetButton(string label, VoicePreset preset, TTSBacke
             var otherPreset = voiceConfig.VoicePresets.First(p => p.Id != preset.Id);
             config.SetCurrentVoicePreset(otherPreset.Id);
 
-            voiceConfig.UngenderedVoicePresets[backend].Remove(preset.Id);
-            voiceConfig.MaleVoicePresets[backend].Remove(preset.Id);
-            voiceConfig.FemaleVoicePresets[backend].Remove(preset.Id);
+            // Use TryGetValue to safely access the inner dictionary for the specific backend
+            if (voiceConfig.UngenderedVoicePresets.TryGetValue(backend, out var ungendered))
+            {
+                ungendered.Remove(preset.Id);
+            }
+
+            if (voiceConfig.MaleVoicePresets.TryGetValue(backend, out var male))
+            {
+                male.Remove(preset.Id);
+            }
+
+            if (voiceConfig.FemaleVoicePresets.TryGetValue(backend, out var female))
+            {
+                female.Remove(preset.Id);
+            }
 
             voiceConfig.VoicePresets.Remove(preset);
 

src/TextToTalk/Backends/ConfigUIDelegates.cs

@@ -6,9 +6,16 @@ public class ConfigUIDelegates : IConfigUIDelegates
     {
         public Action? OpenVoiceUnlockerAction { get; init; }
 
+        public Action? OpenVoiceStylesWindow { get; init; }
+
         public void OpenVoiceUnlocker()
         {
             OpenVoiceUnlockerAction?.Invoke();
         }
+
+        public void OpenVoiceStylesConfig()
+        {
+            OpenVoiceStylesWindow?.Invoke();
+        }
     }
 }

src/TextToTalk/Backends/ElevenLabs/ElevenLabsBackend.cs

@@ -1,4 +1,5 @@
 using System;
+using System.Linq;
 using System.Net;
 using System.Net.Http;
 using System.Threading.Tasks;
@@ -11,12 +12,19 @@ public class ElevenLabsBackend : VoiceBackend
     private readonly ElevenLabsBackendUI ui;
     private readonly ElevenLabsBackendUIModel uiModel;
     private readonly INotificationService notificationService;
+    private readonly PluginConfiguration config;
 
     public ElevenLabsBackend(PluginConfiguration config, HttpClient http, INotificationService notificationService)
     {
         this.uiModel = new ElevenLabsBackendUIModel(config, http);
         this.ui = new ElevenLabsBackendUI(uiModel, config, this);
         this.notificationService = notificationService;
+        this.config = config;
+    }
+
+    public override void DrawStyles(IConfigUIDelegates helpers)
+    {
+        helpers.OpenVoiceStylesConfig();
     }
 
     public override void Say(SayRequest request)
@@ -32,7 +40,7 @@ public override void Say(SayRequest request)
             {
                 await this.uiModel.ElevenLabs.Say(elevenLabsVoicePreset.VoiceId, elevenLabsVoicePreset.PlaybackRate,
                     elevenLabsVoicePreset.Volume, elevenLabsVoicePreset.SimilarityBoost,
-                    elevenLabsVoicePreset.Stability, request.Source, request.Text);
+                    elevenLabsVoicePreset.Stability, request.Source, request.Text, elevenLabsVoicePreset.ModelId);
                 this.uiModel.UpdateUserSubscriptionInfo();
             }
             catch (ElevenLabsUnauthorizedException e)