Added audio and video transcription for chats and assistants (#856)

app/MindWork AI Studio/Tools/Rust/CreateMediaJobRequest.cs

@@ -0,0 +1,7 @@
+namespace AIStudio.Tools.Rust;
+
+/// <summary>Request body used to start a Rust media normalization job.</summary>
+/// <param name="InputPath">Absolute source media path.</param>
+/// <param name="OutputPath">Absolute operation-owned output path.</param>
+/// <param name="MaxPassThroughBytes">Optional pass-through size ceiling.</param>
+public sealed record CreateMediaJobRequest(string InputPath, string OutputPath, ulong? MaxPassThroughBytes = null);

app/MindWork AI Studio/Tools/Rust/CreateMediaJobResponse.cs

@@ -0,0 +1,5 @@
+namespace AIStudio.Tools.Rust;
+
+/// <summary>Response returned after a Rust media job is registered.</summary>
+/// <param name="JobId">Opaque runtime job identifier.</param>
+public sealed record CreateMediaJobResponse(string JobId);

app/MindWork AI Studio/Tools/Rust/FileTypes.cs

@@ -61,7 +61,7 @@ public static class FileTypes
     public static readonly FileTypeFilter IMAGE = FileTypeFilter.Leaf(TB("Image"),
         "jpg", "jpeg", "png", "gif", "bmp", "tiff", "svg", "webp", "heic");
     public static readonly FileTypeFilter AUDIO = FileTypeFilter.Leaf(TB("Audio"),
-        "mp3", "wav", "wave", "aac", "flac", "ogg", "m4a", "wma", "alac", "aiff", "m4b");
+        "mp3", "wav", "wave", "aac", "flac", "ogg", "opus", "m4a", "m4b", "wma", "alac", "aif", "aiff", "caf");
     public static readonly FileTypeFilter VIDEO = FileTypeFilter.Leaf(TB("Video"),
         "mp4", "m4v", "avi", "mkv", "mov", "wmv", "flv", "webm");
 

app/MindWork AI Studio/Tools/Rust/MediaJobError.cs

@@ -0,0 +1,8 @@
+namespace AIStudio.Tools.Rust;
+
+/// <summary>
+/// Runtime media error containing a stable code and an English log diagnostic.
+/// </summary>
+/// <param name="Code">Stable machine-readable error category.</param>
+/// <param name="Message">US-English diagnostic intended for logs.</param>
+public sealed record MediaJobError(MediaJobErrorCode Code, string Message);

app/MindWork AI Studio/Tools/Rust/MediaJobErrorCode.cs

@@ -0,0 +1,85 @@
+namespace AIStudio.Tools.Rust;
+
+/// <summary>
+/// Stable failure categories returned by the Rust media pipeline.
+/// </summary>
+public enum MediaJobErrorCode
+{
+    /// <summary>The runtime returned an unrecognized code.</summary>
+    UNKNOWN,
+    
+    /// <summary>The input file does not exist.</summary>
+    FILE_NOT_FOUND,
+    
+    /// <summary>The file type could not be identified.</summary>
+    UNKNOWN_FORMAT,
+    
+    /// <summary>Executable input was rejected.</summary>
+    UNSAFE_FILE,
+    
+    /// <summary>The input is not media.</summary>
+    NOT_MEDIA,
+    
+    /// <summary>The input file could not be opened.</summary>
+    FILE_OPEN_FAILED,
+    
+    /// <summary>The container is unsupported.</summary>
+    UNSUPPORTED_CONTAINER,
+    
+    /// <summary>The media has no audio track.</summary>
+    NO_AUDIO_TRACK,
+    
+    /// <summary>No audio track has a supported decoder.</summary>
+    UNSUPPORTED_CODEC,
+    
+    /// <summary>Decoded audio parameters are absent or inconsistent.</summary>
+    INVALID_AUDIO_PARAMETERS,
+    
+    /// <summary>The Opus identification header is invalid.</summary>
+    INVALID_OPUS_HEADER,
+    
+    /// <summary>The Opus mapping requires unsupported multistream decoding.</summary>
+    UNSUPPORTED_OPUS_MAPPING,
+    
+    /// <summary>The decoder could not be initialized.</summary>
+    DECODER_INIT_FAILED,
+    
+    /// <summary>The encoder could not be initialized.</summary>
+    ENCODER_INIT_FAILED,
+    
+    /// <summary>The stream changed unexpectedly.</summary>
+    STREAM_RESET,
+    
+    /// <summary>The container is damaged.</summary>
+    DAMAGED_CONTAINER,
+    
+    /// <summary>Audio decoding failed.</summary>
+    DECODE_FAILED,
+    
+    /// <summary>Audio resampling failed.</summary>
+    RESAMPLE_FAILED,
+    
+    /// <summary>Opus encoding failed.</summary>
+    ENCODE_FAILED,
+    
+    /// <summary>The output directory or file could not be created.</summary>
+    OUTPUT_CREATE_FAILED,
+    
+    /// <summary>The output could not be written.</summary>
+    OUTPUT_WRITE_FAILED,
+    
+    /// <summary>The partial output could not be committed.</summary>
+    OUTPUT_COMMIT_FAILED,
+    
+    /// <summary>A WebM relative timestamp overflowed.</summary>
+    WEBM_TIMESTAMP_OVERFLOW,
+    
+    /// <summary>WebM serialization failed.</summary>
+    WEBM_WRITE_FAILED,
+    
+    /// <summary>The job was cancelled.</summary>
+    CANCELLED,
+    
+    /// <summary>The runtime worker failed unexpectedly.</summary>
+    INTERNAL_ERROR,
+}

app/MindWork AI Studio/Tools/Rust/MediaJobEvent.cs

@@ -0,0 +1,12 @@
+namespace AIStudio.Tools.Rust;
+
+/// <summary>Snapshot emitted by the Rust media job event stream.</summary>
+/// <param name="Phase">Current job phase.</param>
+/// <param name="Progress">Optional progress fraction.</param>
+/// <param name="Result">Completed result.</param>
+/// <param name="Error">Failure diagnostic.</param>
+public sealed record MediaJobEvent(
+    MediaJobPhase Phase,
+    double? Progress,
+    MediaJobResult? Result,
+    MediaJobError? Error);

app/MindWork AI Studio/Tools/Rust/MediaJobPhase.cs

@@ -0,0 +1,23 @@
+namespace AIStudio.Tools.Rust;
+
+/// <summary>Lifecycle phases exposed by the Rust media API.</summary>
+public enum MediaJobPhase
+{
+    /// <summary>An unknown future value received from Rust.</summary>
+    UNKNOWN,
+    
+    /// <summary>The runtime is identifying the input and selecting audio.</summary>
+    PROBING,
+    
+    /// <summary>The runtime is normalizing audio.</summary>
+    TRANSCODING,
+    
+    /// <summary>The output was committed successfully.</summary>
+    COMPLETED,
+    
+    /// <summary>The job failed.</summary>
+    FAILED,
+    
+    /// <summary>Cancellation and temporary-output cleanup completed.</summary>
+    CANCELLED,
+}

app/MindWork AI Studio/Tools/Rust/MediaJobResult.cs

@@ -0,0 +1,20 @@
+namespace AIStudio.Tools.Rust;
+
+/// <summary>Successful terminal result returned by Rust media normalization.</summary>
+/// <param name="OutputPath">Committed normalized output path.</param>
+/// <param name="OutputFormat">Stable normalized container used for provider uploads.</param>
+/// <param name="OutputCodec">Stable normalized audio codec used for provider uploads.</param>
+/// <param name="DetectedFormat">Detected container diagnostic.</param>
+/// <param name="DetectedCodec">Selected codec diagnostic.</param>
+/// <param name="DurationMs">Normalized duration in milliseconds.</param>
+/// <param name="PassThrough">Whether the source was copied unchanged.</param>
+/// <param name="HasAudibleSignal">Whether the normalized audio exceeds the practical-silence threshold.</param>
+public sealed record MediaJobResult(
+    string OutputPath,
+    string OutputFormat,
+    string OutputCodec,
+    string DetectedFormat,
+    string DetectedCodec,
+    ulong DurationMs,
+    bool PassThrough,
+    bool HasAudibleSignal);