Added transcription of audio and video files to local data sources (#1081)

Co-authored-by: Paul Koudelka <106623909+PaulKoudelka@users.noreply.github.com>

app/MindWork AI Studio/Tools/Validation/CloudProcessing.cs

@@ -0,0 +1,23 @@
+namespace AIStudio.Tools.Validation;
+
+/// <summary>
+/// What a data source sends to a provider which is not self-hosted.
+/// </summary>
+[Flags]
+public enum CloudProcessing
+{
+    /// <summary>
+    /// Nothing leaves the computer or the organization.
+    /// </summary>
+    NONE = 0,
+
+    /// <summary>
+    /// The content goes to an embedding provider in the cloud.
+    /// </summary>
+    EMBEDDING = 1,
+
+    /// <summary>
+    /// Audio and video files go to a transcription provider in the cloud.
+    /// </summary>
+    TRANSCRIPTION = 2,
+}

app/MindWork AI Studio/Tools/Validation/CloudProcessingExtensions.cs

@@ -0,0 +1,17 @@
+namespace AIStudio.Tools.Validation;
+
+public static class CloudProcessingExtensions
+{
+    /// <summary>
+    /// Whether what the user acknowledged covers what the data source sends to the cloud now.
+    /// </summary>
+    /// <remarks>
+    /// An acknowledgment holds only for the providers it was given for. When the data source starts
+    /// to send its recordings to a transcription provider in the cloud, say since a new pattern
+    /// selects them, the user has to acknowledge that once more.
+    /// </remarks>
+    /// <param name="acknowledged">What the user acknowledged.</param>
+    /// <param name="current">What the data source sends to the cloud with its current settings.</param>
+    /// <returns>True when nothing is sent that the user did not acknowledge.</returns>
+    public static bool Covers(this CloudProcessing acknowledged, CloudProcessing current) => (current & ~acknowledged) is CloudProcessing.NONE;
+}

app/MindWork AI Studio/Tools/Validation/DataSourceValidation.cs

@@ -31,6 +31,8 @@ public sealed class DataSourceValidation
 
     public Func<bool> GetSelectedCloudEmbedding { get; init; } = () => false;
 
+    public Func<bool> GetSelectedCloudTranscription { get; init; } = () => false;
+
     public Func<EmbeddingProvider?> GetSelectedEmbeddingProvider { get; init; } = () => null;
 
     public Func<ConfidenceLevel> GetConfidenceLevel { get; init; } = () => ConfidenceLevel.NONE;
@@ -158,8 +160,8 @@ public sealed class DataSourceValidation
         if(!File.Exists(filePath))
             return TB("The file does not exist. Please select a valid file.");
 
-        if(!FileTypes.IsAllowedPath(filePath, FileTypes.DOCUMENT))
-            return TB("The selected file is not a supported document or readable text file.");
+        if(!FileTypes.IsAllowedPath(filePath, FileTypes.DOCUMENT, FileTypes.AUDIO, FileTypes.VIDEO))
+            return TB("The selected file is not a supported document, readable text file, audio file, or video file.");
         
         return null;
     }
@@ -250,6 +252,9 @@ public sealed class DataSourceValidation
 
     public string? ValidateUserAcknowledgedCloudEmbedding(bool value)
     {
+        if(this.GetSelectedCloudTranscription() && !value)
+            return TB("Please acknowledge that you are aware of the cloud transcription implications.");
+
         if(this.GetSelectedCloudEmbedding() && !value)
             return TB("Please acknowledge that you are aware of the cloud embedding implications.");