Updated whisper.cpp to 1.5.0 (Macoron#60)

* Updated code and windows * Update build_cpp.yml * Update build_cpp.yml * Update build_cpp.yml * Update build_cpp.yml * Update build_cpp.yml * Update build_cpp.yml * Update build_cpp.yml * Update build_cpp.yml * Update build_cpp.yml * Update build_cpp.yml * Updated Linux lib * Disabled metal for now * Updated mac, ios and android * Update README.md * Update test.yml * Update test.yml * Delete .github/workflows/activation.yml * Reset tests * Update test.yml
GameWorkstore · Nov 19, 2023 · f2f432e · f2f432e
1 parent 7f4c33f
commit f2f432e
Showing 6 changed files with 20 additions and 4 deletions.
diff --git a/Plugins/Android/libwhisper.a b/Plugins/Android/libwhisper.a
diff --git a/Plugins/Linux/libwhisper.so b/Plugins/Linux/libwhisper.so
diff --git a/Plugins/MacOS/libwhisper.dylib b/Plugins/MacOS/libwhisper.dylib
diff --git a/Plugins/Windows/libwhisper.dll b/Plugins/Windows/libwhisper.dll
diff --git a/Plugins/iOS/libwhisper.a b/Plugins/iOS/libwhisper.a
diff --git a/Runtime/Native/WhisperNativeParams.cs b/Runtime/Native/WhisperNativeParams.cs
@@ -8,6 +8,7 @@
 using whisper_context_ptr = System.IntPtr;
 using whisper_state_ptr = System.IntPtr;
 using whisper_token = System.Int32;
+using System;
 
 namespace Whisper.Native
 {
@@ -19,11 +20,11 @@ public enum WhisperSamplingStrategy
 
     [UnmanagedFunctionPointer(CallingConvention.StdCall)]
     public delegate void whisper_new_segment_callback(whisper_context_ptr ctx, whisper_state_ptr state,
-        int n_new, System.IntPtr user_data);
+        int n_new, IntPtr user_data);
 
     [UnmanagedFunctionPointer(CallingConvention.StdCall)]
     public delegate void whisper_progress_callback(whisper_context_ptr ctx, whisper_state_ptr state,
-        int progress, System.IntPtr user_data);
+        int progress, IntPtr user_data);
 
     /// <summary>
     /// This is direct copy of C++ struct.
@@ -66,6 +67,7 @@ public unsafe struct WhisperNativeParams
 
         [MarshalAs(UnmanagedType.U1)] public bool translate;
         [MarshalAs(UnmanagedType.U1)] public bool no_context; // do not use past transcription (if any) as initial prompt for the decoder
+        [MarshalAs(UnmanagedType.U1)] bool no_timestamps;     // do not generate timestamps
         [MarshalAs(UnmanagedType.U1)] public bool single_segment; // force single segment output (useful for streaming)
         [MarshalAs(UnmanagedType.U1)] public bool print_special; // print special tokens (e.g. <SOT>, <EOT>, <BEG>, etc.)
         [MarshalAs(UnmanagedType.U1)] public bool print_progress; // print progress information
@@ -83,8 +85,12 @@ public unsafe struct WhisperNativeParams
         // [EXPERIMENTAL] speed-up techniques
         // note: these can significantly reduce the quality of the output
         [MarshalAs(UnmanagedType.U1)] public bool speed_up; // speed-up the audio by 2x using Phase Vocoder
+        [MarshalAs(UnmanagedType.U1)] bool debug_mode;        // enable debug_mode provides extra info (eg. Dump log_mel)
         public int audio_ctx; // overwrite the audio context size (0 = use default)
 
+        // [EXPERIMENTAL] [TDRZ] tinydiarize
+        [MarshalAs(UnmanagedType.U1)] bool tdrz_enable;       // enable tinydiarize speaker turn detection
+
         // tokens to provide to the whisper decoder as initial prompt
         // these are prepended to any existing text context from a previous call
         public byte* initial_prompt;
@@ -129,18 +135,28 @@ struct beam_search_struct
 
         // called for every newly generated text segment
         public whisper_new_segment_callback new_segment_callback;
-        public System.IntPtr new_segment_callback_user_data;
+        public IntPtr new_segment_callback_user_data;
 
         // called on each progress update
         public whisper_progress_callback progress_callback;
-        public System.IntPtr progress_callback_user_data;
+        public IntPtr progress_callback_user_data;
 
         // called each time before the encoder starts
         void* encoder_begin_callback;
         void* encoder_begin_callback_user_data;
 
+        // called each time before ggml computation starts
+        void* abort_callback;
+        void* abort_callback_user_data;
+
         // called by each decoder to filter obtained logits
         void* logits_filter_callback;
         void* logits_filter_callback_user_data;
+
+
+        IntPtr grammar_rules;
+        UIntPtr n_grammar_rules;
+        UIntPtr i_start_rule;
+        float grammar_penalty;
     }
 }