Implemented the transcription API (#623)

app/MindWork AI Studio/Components/SecretInputField.razor

@@ -15,6 +15,6 @@
         UserAttributes="@SPELLCHECK_ATTRIBUTES"/>
 
     <MudTooltip Text="@this.ToggleVisibilityTooltip">
-        <MudIconButton Icon="@this.InputTypeIcon" OnClick="() => this.ToggleVisibility()"/>
+        <MudIconButton Icon="@this.InputTypeIcon" OnClick="@(() => this.ToggleVisibility())"/>
     </MudTooltip>
 </MudStack>

app/MindWork AI Studio/Components/Settings/SettingsPanelEmbeddings.razor.cs

@@ -100,13 +100,13 @@ public partial class SettingsPanelEmbeddings : SettingsPanelBase
         if (dialogResult is null || dialogResult.Canceled)
             return;
         
-        var deleteSecretResponse = await this.RustService.DeleteAPIKey(provider);
+        var deleteSecretResponse = await this.RustService.DeleteAPIKey(provider, SecretStoreType.EMBEDDING_PROVIDER);
         if(deleteSecretResponse.Success)
         {
             this.SettingsManager.ConfigurationData.EmbeddingProviders.Remove(provider);
             await this.SettingsManager.StoreSettings();
         }
-        
+
         await this.UpdateEmbeddingProviders();
         await this.MessageBus.SendMessage<bool>(this, Event.CONFIGURATION_CHANGED);
     }

app/MindWork AI Studio/Components/Settings/SettingsPanelProviders.razor.cs

@@ -107,7 +107,7 @@ public partial class SettingsPanelProviders : SettingsPanelBase
         if (dialogResult is null || dialogResult.Canceled)
             return;
         
-        var deleteSecretResponse = await this.RustService.DeleteAPIKey(provider);
+        var deleteSecretResponse = await this.RustService.DeleteAPIKey(provider, SecretStoreType.LLM_PROVIDER);
         if(deleteSecretResponse.Success)
         {
             this.SettingsManager.ConfigurationData.Providers.Remove(provider);

app/MindWork AI Studio/Components/Settings/SettingsPanelTranscription.razor

@@ -10,7 +10,7 @@
             @T("Configured Transcription Providers")
         </MudText>
         <MudJustifiedText Typo="Typo.body1" Class="mb-3">
-            @T("With the support of transcription models, MindWork AI Studio can convert human speech into text. This is useful, for example, when you need to dictate text. You can choose from dedicated transcription models, but not multimodal LLMs (large language models) that can handle both speech and text. The configuration of multimodal models is done in the \"Configure providers\" section.")
+            @T("With the support of transcription models, MindWork AI Studio can convert human speech into text. This is useful, for example, when you need to dictate text. You can choose from dedicated transcription models, but not multimodal LLMs (large language models) that can handle both speech and text. The configuration of multimodal models is done in the 'Configure providers' section.")
         </MudJustifiedText>
         
         <MudTable Items="@this.SettingsManager.ConfigurationData.TranscriptionProviders" Hover="@true" Class="border-dashed border rounded-lg">

app/MindWork AI Studio/Components/Settings/SettingsPanelTranscription.razor.cs

@@ -100,13 +100,13 @@ public partial class SettingsPanelTranscription : SettingsPanelBase
         if (dialogResult is null || dialogResult.Canceled)
             return;
         
-        var deleteSecretResponse = await this.RustService.DeleteAPIKey(provider);
+        var deleteSecretResponse = await this.RustService.DeleteAPIKey(provider, SecretStoreType.TRANSCRIPTION_PROVIDER);
         if(deleteSecretResponse.Success)
         {
             this.SettingsManager.ConfigurationData.TranscriptionProviders.Remove(provider);
             await this.SettingsManager.StoreSettings();
         }
-        
+
         await this.UpdateTranscriptionProviders();
         await this.MessageBus.SendMessage<bool>(this, Event.CONFIGURATION_CHANGED);
     }

app/MindWork AI Studio/Components/VoiceRecorder.razor

@@ -6,11 +6,18 @@
 @if (PreviewFeatures.PRE_SPEECH_TO_TEXT_2026.IsEnabled(this.SettingsManager) && !string.IsNullOrWhiteSpace(this.SettingsManager.ConfigurationData.App.UseTranscriptionProvider))
 {
     <MudTooltip Text="@this.Tooltip">
-        <MudToggleIconButton Toggled="@this.isRecording"
-                             ToggledChanged="@this.OnRecordingToggled"
-                             Icon="@Icons.Material.Filled.Mic"
-                             ToggledIcon="@Icons.Material.Filled.Stop"
-                             Color="Color.Primary"
-                             ToggledColor="Color.Error"/>
+        @if (this.isTranscribing)
+        {
+            <MudProgressCircular Size="Size.Small" Indeterminate="true" Color="Color.Primary"/>
+        }
+        else
+        {
+            <MudToggleIconButton Toggled="@this.isRecording"
+                                 ToggledChanged="@this.OnRecordingToggled"
+                                 Icon="@Icons.Material.Filled.Mic"
+                                 ToggledIcon="@Icons.Material.Filled.Stop"
+                                 Color="Color.Primary"
+                                 ToggledColor="Color.Error"/>
+        }
     </MudTooltip>
 }

app/MindWork AI Studio/Components/VoiceRecorder.razor.cs

@@ -1,3 +1,4 @@
+using AIStudio.Provider;
 using AIStudio.Tools.MIME;
 using AIStudio.Tools.Services;
 
@@ -9,21 +10,30 @@ public partial class VoiceRecorder : MSGComponentBase
 {
     [Inject]
     private ILogger<VoiceRecorder> Logger { get; init; } = null!;
-    
+
     [Inject]
     private IJSRuntime JsRuntime { get; init; } = null!;
-    
+
     [Inject]
     private RustService RustService { get; init; } = null!;
 
+    [Inject]
+    private ISnackbar Snackbar { get; init; } = null!;
+
     private uint numReceivedChunks;
     private bool isRecording;
+    private bool isTranscribing;
     private FileStream? currentRecordingStream;
     private string? currentRecordingPath;
     private string? currentRecordingMimeType;
+    private string? finalRecordingPath;
     private DotNetObjectReference<VoiceRecorder>? dotNetReference;
-    
-    private string Tooltip => this.isRecording ? T("Stop recording and start transcription") : T("Start recording your voice for a transcription");
+
+    private string Tooltip => this.isTranscribing
+        ? T("Transcription in progress...")
+        : this.isRecording
+            ? T("Stop recording and start transcription")
+            : T("Start recording your voice for a transcription");
     
     private async Task OnRecordingToggled(bool toggled)
     {
@@ -66,6 +76,10 @@ public partial class VoiceRecorder : MSGComponentBase
 
             this.isRecording = false;
             this.StateHasChanged();
+
+            // Start transcription if we have a recording and a configured provider:
+            if (this.finalRecordingPath is not null)
+                await this.TranscribeRecordingAsync();
         }
     }
 
@@ -127,6 +141,7 @@ public partial class VoiceRecorder : MSGComponentBase
 
     private async Task FinalizeRecordingStream()
     {
+        this.finalRecordingPath = null;
         if (this.currentRecordingStream is not null)
         {
             await this.currentRecordingStream.FlushAsync();
@@ -142,6 +157,7 @@ public partial class VoiceRecorder : MSGComponentBase
                 if (File.Exists(this.currentRecordingPath))
                 {
                     File.Move(this.currentRecordingPath, newPath, overwrite: true);
+                    this.finalRecordingPath = newPath;
                     this.Logger.LogInformation("Finalized audio recording over {NumChunks} streamed audio chunks to the file '{RecordingPath}'.", this.numReceivedChunks, newPath);
                 }
             }
@@ -170,6 +186,114 @@ public partial class VoiceRecorder : MSGComponentBase
         };
     }
 
+    private async Task TranscribeRecordingAsync()
+    {
+        if (this.finalRecordingPath is null)
+            return;
+
+        this.isTranscribing = true;
+        this.StateHasChanged();
+
+        try
+        {
+            // Get the configured transcription provider ID:
+            var transcriptionProviderId = this.SettingsManager.ConfigurationData.App.UseTranscriptionProvider;
+            if (string.IsNullOrWhiteSpace(transcriptionProviderId))
+            {
+                this.Logger.LogWarning("No transcription provider is configured.");
+                await this.MessageBus.SendError(new(Icons.Material.Filled.VoiceChat, this.T("No transcription provider is configured.")));
+                return;
+            }
+
+            // Find the transcription provider in the list of configured providers:
+            var transcriptionProviderSettings = this.SettingsManager.ConfigurationData.TranscriptionProviders
+                .FirstOrDefault(x => x.Id == transcriptionProviderId);
+
+            if (transcriptionProviderSettings is null)
+            {
+                this.Logger.LogWarning("The configured transcription provider with ID '{ProviderId}' was not found.", transcriptionProviderId);
+                await this.MessageBus.SendError(new(Icons.Material.Filled.VoiceChat, this.T("The configured transcription provider was not found.")));
+                return;
+            }
+
+            // Check the confidence level:
+            var minimumLevel = this.SettingsManager.GetMinimumConfidenceLevel(Tools.Components.NONE);
+            var providerConfidence = transcriptionProviderSettings.UsedLLMProvider.GetConfidence(this.SettingsManager);
+            if (providerConfidence.Level < minimumLevel)
+            {
+                this.Logger.LogWarning(
+                    "The configured transcription provider '{ProviderName}' has a confidence level of '{ProviderLevel}', which is below the minimum required level of '{MinimumLevel}'.",
+                    transcriptionProviderSettings.Name,
+                    providerConfidence.Level,
+                    minimumLevel);
+                await this.MessageBus.SendError(new(Icons.Material.Filled.VoiceChat, this.T("The configured transcription provider does not meet the minimum confidence level.")));
+                return;
+            }
+
+            // Create the provider instance:
+            var provider = transcriptionProviderSettings.CreateProvider();
+            if (provider.Provider is LLMProviders.NONE)
+            {
+                this.Logger.LogError("Failed to create the transcription provider instance.");
+                await this.MessageBus.SendError(new(Icons.Material.Filled.VoiceChat, this.T("Failed to create the transcription provider.")));
+                return;
+            }
+
+            // Call the transcription API:
+            this.Logger.LogInformation("Starting transcription with provider '{ProviderName}' and model '{ModelName}'.", transcriptionProviderSettings.Name, transcriptionProviderSettings.Model.DisplayName);
+            var transcribedText = await provider.TranscribeAudioAsync(transcriptionProviderSettings.Model, this.finalRecordingPath, this.SettingsManager);
+
+            if (string.IsNullOrWhiteSpace(transcribedText))
+            {
+                this.Logger.LogWarning("The transcription result is empty.");
+                await this.MessageBus.SendWarning(new(Icons.Material.Filled.VoiceChat, this.T("The transcription result is empty.")));
+                return;
+            }
+            
+            // Remove trailing and leading whitespace:
+            transcribedText = transcribedText.Trim();
+            
+            // Replace line breaks with spaces:
+            transcribedText = transcribedText.Replace("\r", " ").Replace("\n", " ");
+            
+            // Replace two spaces with a single space:
+            transcribedText = transcribedText.Replace("  ", " ");
+
+            this.Logger.LogInformation("Transcription completed successfully. Result length: {Length} characters.", transcribedText.Length);
+
+            // Play the transcription done sound effect:
+            await this.JsRuntime.InvokeVoidAsync("playSound", "/sounds/transcription_done.ogg");
+
+            // Copy the transcribed text to the clipboard:
+            await this.RustService.CopyText2Clipboard(this.Snackbar, transcribedText);
+
+            // Delete the recording file:
+            try
+            {
+                if (File.Exists(this.finalRecordingPath))
+                {
+                    File.Delete(this.finalRecordingPath);
+                    this.Logger.LogInformation("Deleted the recording file '{RecordingPath}'.", this.finalRecordingPath);
+                }
+            }
+            catch (Exception ex)
+            {
+                this.Logger.LogError(ex, "Failed to delete the recording file '{RecordingPath}'.", this.finalRecordingPath);
+            }
+        }
+        catch (Exception ex)
+        {
+            this.Logger.LogError(ex, "An error occurred during transcription.");
+            await this.MessageBus.SendError(new(Icons.Material.Filled.VoiceChat, this.T("An error occurred during transcription.")));
+        }
+        finally
+        {
+            this.finalRecordingPath = null;
+            this.isTranscribing = false;
+            this.StateHasChanged();
+        }
+    }
+
     private sealed class AudioRecordingResult
     {
         public string MimeType { get; init; } = string.Empty;