Updated whisper.cpp to 1.5.0 (#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

.github/workflows/build_cpp.yml

@@ -9,7 +9,7 @@ on:
  whisper_cpp_repo_ref:
  description: 'Tag, branch or commit'
  required: true
- default: 'v1.4.2'
+ default: 'v1.5.0'
 jobs:
  build-windows:
  name: Build for Windows (x86_64)
@@ -61,14 +61,24 @@ jobs:
  ref: ${{ github.event.inputs.whisper_cpp_repo_ref }}
  path: whisper-cpp
 
- - name: Dependencies
+ - name: Latest Cmake
  run: |
- apt-get update
- apt-get install -y build-essential
- apt-get install -y cmake
+ apt-get update \
+ && apt-get install -y build-essential \
+ && apt-get install -y wget \
+ && rm -rf /var/lib/apt/lists/* \
+ && wget https://github.com/Kitware/CMake/releases/download/v3.24.1/cmake-3.24.1-Linux-x86_64.sh \
+ -q -O /tmp/cmake-install.sh \
+ && chmod u+x /tmp/cmake-install.sh \
+ && mkdir /opt/cmake-3.24.1 \
+ && /tmp/cmake-install.sh --skip-license --prefix=/opt/cmake-3.24.1 \
+ && rm /tmp/cmake-install.sh \
+ && ln -s /opt/cmake-3.24.1/bin/* /usr/local/bin
 
  - name: Change gcc and g++
- run: apt-get install -y gcc-8 g++-8
+ run: |
+ apt-get update
+ apt-get install -y gcc-8 g++-8
  
  - name: Run build script
  run: |

.github/workflows/test.yml

@@ -19,9 +19,6 @@ jobs:
  matrix:
  unityVersion:
  - 2019.4.40f1
- - 2020.3.48f1
- - 2021.3.25f1
- - 2022.2.19f1
  steps:
  - uses: actions/checkout@v3
  - uses: game-ci/unity-test-runner@v2

Packages/com.whisper.unity/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;
  }
 }

README.md

@@ -1,5 +1,5 @@
 # whisper.unity
-[![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg)](https://opensource.org/licenses/MIT) [![whisper.cpp](https://img.shields.io/badge/whisper.cpp-v1.4.2-green)](https://github.com/ggerganov/whisper.cpp/releases/tag/v1.4.2)
+[![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg)](https://opensource.org/licenses/MIT) [![whisper.cpp](https://img.shields.io/badge/whisper.cpp-v1.5.0-green)](https://github.com/ggerganov/whisper.cpp/releases/tag/v1.5.0)
 
 [![Testing](https://github.com/Macoron/whisper.unity/actions/workflows/test.yml/badge.svg)](https://github.com/Macoron/whisper.unity/actions/workflows/test.yml)
 
@@ -51,7 +51,7 @@ This project comes with prebuild libraries of whisper.cpp for all supported plat
 
 In case you want to build libraries on your machine:
 1. Clone the original [whisper.cpp](https://github.com/ggerganov/whisper.cpp) repository
-2. Checkout tag [v1.4.2](https://github.com/ggerganov/whisper.cpp/releases/tag/v1.4.2). Other versions might not work with this Unity bindings.
+2. Checkout tag [v1.5.0](https://github.com/ggerganov/whisper.cpp/tree/v1.5.0). Other versions might not work with this Unity bindings.
 3. Open whisper.unity folder with command line
 4. If you are using **Windows** write:
 ```bash

build_cpp.sh

@@ -16,8 +16,9 @@ build_mac() {
  clean_build
  echo "Starting building for Mac..."
 
- cmake -DCMAKE_OSX_ARCHITECTURES="x86_64;arm64" -DWHISPER_BUILD_TESTS=OFF -DWHISPER_BUILD_EXAMPLES=OFF \
- -DWHISPER_NO_AVX=ON -DWHISPER_NO_AVX2=ON -DWHISPER_NO_FMA=ON -DWHISPER_NO_F16C=ON ../
+ cmake -DCMAKE_OSX_ARCHITECTURES="x86_64;arm64" -DWHISPER_METAL=OFF \
+ -DWHISPER_NO_AVX=ON -DWHISPER_NO_AVX2=ON -DWHISPER_NO_FMA=ON -DWHISPER_NO_F16C=ON \
+ -DWHISPER_BUILD_TESTS=OFF -DWHISPER_BUILD_EXAMPLES=OFF ../
  make
 
  echo "Build for Mac complete!"
@@ -34,7 +35,7 @@ build_ios() {
  echo "Starting building for ios..."
 
  cmake -DBUILD_SHARED_LIBS=OFF -DCMAKE_SYSTEM_NAME=iOS \
- -DCMAKE_OSX_ARCHITECTURES="arm64;x86_64" \
+ -DCMAKE_OSX_ARCHITECTURES="x86_64;arm64" -DWHISPER_METAL=OFF \
  -DCMAKE_SYSTEM_PROCESSOR=arm64 -DCMAKE_IOS_INSTALL_COMBINED=YES \
  -DWHISPER_BUILD_TESTS=OFF -DWHISPER_BUILD_EXAMPLES=OFF ../
  make