ruby : add VAD speech segments API

bindings/ruby/README.md

@@ -117,7 +117,7 @@ Whisper::Params.new(
     threshold: 1.0, # defaults to 0.5
     min_speech_duration_ms: 500, # defaults to 250
     min_silence_duration_ms: 200, # defaults to 100
-    max_speech_duration_s: 30000, # default is FLT_MAX,
+    max_speech_duration_s: 30000.0, # default is FLT_MAX,
     speech_pad_ms: 50, # defaults to 30
     samples_overlap: 0.5 # defaults to 0.1
   ),
@@ -176,15 +176,15 @@ See whisper.cpp's [README](https://github.com/ggml-org/whisper.cpp/blob/master/R
 Boolean options:
 
 * `-DGGML_BLAS=1` -> `--enable-ggml-blas`
-* `-DWHISER_COREML=OFF` -> `--disable-whisper-coreml`
+* `-DWHISPER_COREML=OFF` -> `--disable-whisper-coreml`
 
 Argument options:
 
 * `-DGGML_CUDA_COMPRESSION_MODE=size` -> `--ggml-cuda-compression-mode=size`
 
 Combination:
 
-* `-DGGML_CUDA=1 -DCMAKE_CUDA_ARCHITECTURES="86"` -> `--enable-ggml-cuda --cmake_cuda-architectures="86"`
+* `-DGGML_CUDA=1 -DCMAKE_CUDA_ARCHITECTURES="86"` -> `--enable-ggml-cuda --cmake-cuda-architectures="86"`
 
 For boolean options like `GGML_CUDA`, the README says `-DGGML_CUDA=1`. You need strip `-D`, prepend `--enable-` for `1` or `ON` (`--disable-` for `0` or `OFF`) and make it kebab-case: `--enable-ggml-cuda`.  
 For options which require arguments like `CMAKE_CUDA_ARCHITECTURES`, the README says `-DCMAKE_CUDA_ARCHITECTURES="86"`. You need strip `-D`, prepend `--`, make it kebab-case, append `=` and append argument: `--cmake-cuda-architectures="86"`.
@@ -478,13 +478,14 @@ Development
     % cd whisper.cpp/bindings/ruby
     % rake test
 
-First call of `rake test` builds an extension and downloads a model for testing. After that, you add tests in `tests` directory and modify `ext/ruby_whisper.cpp`.
+First call of `rake test` builds an extension and downloads a model for testing. After that, you add tests in `test` directory and modify `ext/*.{h,c,cpp}`.
 
 If something seems wrong on build, running `rake clean` solves some cases.
 
 ### Need help ###
 
 * Windows support
+* Check of compilation with various hardware
 * Refinement of C/C++ code, especially memory management
 
 License

bindings/ruby/ext/options.rb

@@ -105,7 +105,7 @@ def write_cache_file
     FileUtils.mkpath File.dirname(cache_path)
     File.open cache_path, "w" do |file|
       @options.reject {|name, (type, value)| value.nil?}.each do |name, (type, value)|
-        line = "set(CACHE{%<name>s} TYPE %<type>s FORCE VALUE %<value>s)" % {
+        line = 'set(%<name>s %<value>s CACHE %<type>s "" FORCE)' %{
           name:,
           type:,
           value: value == true ? "ON" : value == false ? "OFF" : escape_cmake(value)

bindings/ruby/ext/ruby_whisper.c

@@ -133,7 +133,6 @@ void Init_whisper() {
   id_coreml_compiled_models = rb_intern("coreml_compiled_models");
   id_cache = rb_intern("cache");
   id_n_processors = rb_intern("n_processors");
-  id_extended = rb_intern("extended");
   id_start_log_callback_thread = rb_intern("start_log_callback_thread");
   id_log_callback_thread = rb_intern("@log_callback_thread");
   id_alive_p = rb_intern("alive?");

bindings/ruby/ext/ruby_whisper.h

@@ -110,6 +110,13 @@ typedef struct {
   int n_samples;
 } ruby_whisper_full_args;
 
+typedef struct segments_from_samples_args {
+  VALUE *context;
+  VALUE *params;
+  float *samples;
+  int n_samples;
+} segments_from_samples_args;
+
 typedef struct ruby_whisper_full_parallel_args {
   VALUE *context;
   VALUE *params;
@@ -149,7 +156,6 @@ typedef struct {
   VALUE context;
 } ruby_whisper_parakeet_model;
 
-extern ID id_extended;
 extern ID id_log_callback_thread;
 extern ID id_start_log_callback_thread;
 extern ID id_alive_p;

bindings/ruby/ext/ruby_whisper_context.c

@@ -349,14 +349,21 @@ fill_samples(VALUE rb_args)
 
   if (RB_TYPE_P(*args->src, T_ARRAY)) {
     for (int i = 0; i < args->n_samples; i++) {
-      args->dest[i] = RFLOAT_VALUE(rb_ary_entry(*args->src, i));
+      VALUE sample = rb_ary_entry(*args->src, i);
+      if (!RB_FLOAT_TYPE_P(sample)) {
+        sample = rb_to_float(sample);
+      }
+      args->dest[i] = RFLOAT_VALUE(sample);
     }
   } else {
     // TODO: use rb_block_call
     VALUE iter = rb_funcall(*args->src, id_to_enum, 1, rb_str_new2("each"));
     for (int i = 0; i < args->n_samples; i++) {
       // TODO: check if iter is exhausted and raise ArgumentError appropriately
       VALUE sample = rb_funcall(iter, id_next, 0);
+      if (!RB_FLOAT_TYPE_P(sample)) {
+        sample = rb_to_float(sample);
+      }
       args->dest[i] = RFLOAT_VALUE(sample);
     }
   }
@@ -745,6 +752,45 @@ ruby_whisper_full_get_segment_no_speech_prob(VALUE self, VALUE i_segment)
   return DBL2NUM(no_speech_prob);
 }
 
+static VALUE
+ruby_whisper_full_n_vad_segments(VALUE self)
+{
+  ruby_whisper *rw;
+  GetContext(self, rw);
+
+  return INT2NUM(whisper_full_n_vad_segments(rw->context));
+}
+
+static int
+ruby_whisper_full_check_vad_segment_index(const ruby_whisper *rw, const VALUE i_segment)
+{
+    const int c_i_segment = NUM2INT(i_segment);
+    if (c_i_segment < 0 || c_i_segment >= whisper_full_n_vad_segments(rw->context)) {
+      rb_raise(rb_eIndexError, "segment index %d out of range", c_i_segment);
+    }
+    return c_i_segment;
+}
+
+static VALUE
+ruby_whisper_full_get_vad_segment_t0(VALUE self, VALUE i_segment)
+{
+  ruby_whisper *rw;
+  GetContext(self, rw);
+  const int c_i_segment = ruby_whisper_full_check_vad_segment_index(rw, i_segment);
+
+  return LONG2NUM(whisper_full_get_vad_segment_t0(rw->context, c_i_segment));
+}
+
+static VALUE
+ruby_whisper_full_get_vad_segment_t1(VALUE self, VALUE i_segment)
+{
+  ruby_whisper *rw;
+  GetContext(self, rw);
+  const int c_i_segment = ruby_whisper_full_check_vad_segment_index(rw, i_segment);
+
+  return LONG2NUM(whisper_full_get_vad_segment_t1(rw->context, c_i_segment));
+}
+
 // High level API
 
 static VALUE
@@ -830,6 +876,9 @@ init_ruby_whisper_context(VALUE *mWhisper)
   rb_define_method(cContext, "full_get_segment_speaker_turn_next", ruby_whisper_full_get_segment_speaker_turn_next, 1);
   rb_define_method(cContext, "full_get_segment_text", ruby_whisper_full_get_segment_text, 1);
   rb_define_method(cContext, "full_get_segment_no_speech_prob", ruby_whisper_full_get_segment_no_speech_prob, 1);
+  rb_define_method(cContext, "full_n_vad_segments", ruby_whisper_full_n_vad_segments, 0);
+  rb_define_method(cContext, "full_get_vad_segment_t0", ruby_whisper_full_get_vad_segment_t0, 1);
+  rb_define_method(cContext, "full_get_vad_segment_t1", ruby_whisper_full_get_vad_segment_t1, 1);
   rb_define_method(cContext, "full", ruby_whisper_full, -1);
   rb_define_method(cContext, "full_parallel", ruby_whisper_full_parallel, -1);
 

bindings/ruby/ext/ruby_whisper_log_settable.h

@@ -41,7 +41,6 @@
   rb_define_singleton_method(mod, "drain_logs", ruby_whisper_##log_queue##_s_drain_logs, 0); \
   rb_define_singleton_method(mod, "log_set", ruby_whisper_##log_queue##_s_log_set, 2); \
   rb_set_end_proc(ruby_whisper_##log_queue##_end_proc, Qnil); \
-  rb_extend_object(mod, mLogSettable); \
-  rb_funcall(mLogSettable, id_extended, 1, mod);
+  rb_extend_object(mod, mLogSettable);
 
 #endif

bindings/ruby/ext/ruby_whisper_params.c

@@ -446,7 +446,7 @@ ruby_whisper_params_allocate(VALUE klass)
     rwp->params.vad_model_path = ruby_strdup(rwp->params.vad_model_path);
   }
   rwp->diarize = false;
-  rwp->vad_params = TypedData_Wrap_Struct(cVADParams, &ruby_whisper_vad_params_type, (void *)&rwp->params.vad_params);
+  rwp->vad_params = rb_class_new_instance(0, NULL, cVADParams);
   rwp->new_segment_callback_container = ruby_whisper_callback_container_allocate();
   rwp->progress_callback_container = ruby_whisper_callback_container_allocate();
   rwp->encoder_begin_callback_container = ruby_whisper_callback_container_allocate();
@@ -918,7 +918,7 @@ ruby_whisper_params_set_temperature(VALUE self, VALUE value)
 {
   ruby_whisper_params *rwp;
   TypedData_Get_Struct(self, ruby_whisper_params, &ruby_whisper_params_type, rwp);
-  rwp->params.temperature = RFLOAT_VALUE(value);
+  rwp->params.temperature = NUM2DBL(value);
   return value;
 }
 /*
@@ -943,7 +943,7 @@ ruby_whisper_params_set_max_initial_ts(VALUE self, VALUE value)
 {
   ruby_whisper_params *rwp;
   TypedData_Get_Struct(self, ruby_whisper_params, &ruby_whisper_params_type, rwp);
-  rwp->params.max_initial_ts = RFLOAT_VALUE(value);
+  rwp->params.max_initial_ts = NUM2DBL(value);
   return value;
 }
 /*
@@ -966,7 +966,7 @@ ruby_whisper_params_set_length_penalty(VALUE self, VALUE value)
 {
   ruby_whisper_params *rwp;
   TypedData_Get_Struct(self, ruby_whisper_params, &ruby_whisper_params_type, rwp);
-  rwp->params.length_penalty = RFLOAT_VALUE(value);
+  rwp->params.length_penalty = NUM2DBL(value);
   return value;
 }
 /*
@@ -989,7 +989,7 @@ ruby_whisper_params_set_temperature_inc(VALUE self, VALUE value)
 {
   ruby_whisper_params *rwp;
   TypedData_Get_Struct(self, ruby_whisper_params, &ruby_whisper_params_type, rwp);
-  rwp->params.temperature_inc = RFLOAT_VALUE(value);
+  rwp->params.temperature_inc = NUM2DBL(value);
   return value;
 }
 /*
@@ -1014,7 +1014,7 @@ ruby_whisper_params_set_entropy_thold(VALUE self, VALUE value)
 {
   ruby_whisper_params *rwp;
   TypedData_Get_Struct(self, ruby_whisper_params, &ruby_whisper_params_type, rwp);
-  rwp->params.entropy_thold = RFLOAT_VALUE(value);
+  rwp->params.entropy_thold = NUM2DBL(value);
   return value;
 }
 /*
@@ -1037,7 +1037,7 @@ ruby_whisper_params_set_logprob_thold(VALUE self, VALUE value)
 {
   ruby_whisper_params *rwp;
   TypedData_Get_Struct(self, ruby_whisper_params, &ruby_whisper_params_type, rwp);
-  rwp->params.logprob_thold = RFLOAT_VALUE(value);
+  rwp->params.logprob_thold = NUM2DBL(value);
   return value;
 }
 /*
@@ -1060,7 +1060,7 @@ ruby_whisper_params_set_no_speech_thold(VALUE self, VALUE value)
 {
   ruby_whisper_params *rwp;
   TypedData_Get_Struct(self, ruby_whisper_params, &ruby_whisper_params_type, rwp);
-  rwp->params.no_speech_thold = RFLOAT_VALUE(value);
+  rwp->params.no_speech_thold = NUM2DBL(value);
   return value;
 }
 static VALUE
@@ -1307,6 +1307,7 @@ static VALUE
 ruby_whisper_params_set_vad_params(VALUE self, VALUE value)
 {
   ruby_whisper_params *rwp;
+  rb_check_typeddata(value, &ruby_whisper_vad_params_type);
   TypedData_Get_Struct(self, ruby_whisper_params, &ruby_whisper_params_type, rwp);
   rwp->vad_params = value;
   return value;

bindings/ruby/ext/ruby_whisper_transcribe.cpp

@@ -15,28 +15,9 @@ extern ID id_call;
 extern ID id_to_path;
 extern ID transcribe_option_names[1];
 
-extern void prepare_transcription(ruby_whisper_params * rwp, VALUE * self, int n_processors);
 extern VALUE full_body(VALUE rb_args);
 extern VALUE full_parallel_body(VALUE rb_args);
 
-typedef struct{
-  struct whisper_context *context;
-  struct whisper_full_params *params;
-  float *samples;
-  size_t n_samples;
-  int n_processors;
-  int result;
-} transcribe_without_gvl_args;
-
-static void*
-transcribe_without_gvl(void *rb_args)
-{
-  transcribe_without_gvl_args *args = (transcribe_without_gvl_args *)rb_args;
-  args->result = whisper_full_parallel(args->context, *args->params, args->samples, args->n_samples, args->n_processors);
-
-  return NULL;
-}
-
 /*
  * transcribe a single file
  * can emit to a block results

bindings/ruby/ext/ruby_whisper_vad_context.c

@@ -12,13 +12,6 @@ extern VALUE release_samples(VALUE parsed);
 
 extern VALUE ruby_whisper_vad_segments_s_init(struct whisper_vad_segments *segments);
 
-typedef struct segments_from_samples_args {
-  VALUE *context;
-  VALUE *params;
-  float *samples;
-  int n_samples;
-} segments_from_samples_args;
-
 static size_t
 ruby_whisper_vad_context_memsize(const void *p)
 {
@@ -77,7 +70,24 @@ ruby_whisper_vad_context_initialize(VALUE self, VALUE model_path)
   return Qnil;
 }
 
-static VALUE
+typedef struct {
+  struct whisper_vad_context *context;
+  struct whisper_vad_params *params;
+  float *samples;
+  int n_samples;
+  struct whisper_vad_segments *result;
+} segments_from_samples_without_gvl_args;
+
+static void*
+segments_from_samples_without_gvl(void *rb_args)
+{
+  segments_from_samples_without_gvl_args *args = (segments_from_samples_without_gvl_args *)rb_args;
+  args->result = whisper_vad_segments_from_samples(args->context, *args->params, args->samples, args->n_samples);
+
+  return NULL;
+}
+
+VALUE
 segments_from_samples_body(VALUE rb_args)
 {
   segments_from_samples_args *args = (segments_from_samples_args *)rb_args;
@@ -87,9 +97,19 @@ segments_from_samples_body(VALUE rb_args)
   GetVADContext(*args->context, rwvc);
   GetVADParams(*args->params, rwvp);
 
-  struct whisper_vad_segments *segments = whisper_vad_segments_from_samples(rwvc->context, rwvp->params, args->samples, args->n_samples);
+  segments_from_samples_without_gvl_args segments_from_samples_without_gvl_args = {
+    rwvc->context,
+    &rwvp->params,
+    args->samples,
+    args->n_samples,
+    NULL
+  };
+  rb_thread_call_without_gvl(segments_from_samples_without_gvl, (void *)&segments_from_samples_without_gvl_args, NULL, NULL);
+  if (!segments_from_samples_without_gvl_args.result) {
+    rb_raise(rb_eRuntimeError, "Failed to process audio");
+  }
 
-  return ruby_whisper_vad_segments_s_init(segments);
+  return ruby_whisper_vad_segments_s_init(segments_from_samples_without_gvl_args.result);
 }
 
 static VALUE

bindings/ruby/ext/ruby_whisper_vad_context_detect.cpp

@@ -16,18 +16,16 @@ extern const rb_data_type_t ruby_whisper_vad_params_type;
 extern const rb_data_type_t ruby_whisper_vad_segments_type;
 
 extern VALUE ruby_whisper_vad_segments_s_init(struct whisper_vad_segments *segments);
+extern VALUE segments_from_samples_body(VALUE rb_args);
 
 VALUE
 ruby_whisper_vad_detect(VALUE self, VALUE file_path, VALUE params) {
   ruby_whisper_vad_context *rwvc;
-  ruby_whisper_vad_params *rwvp;
   std::string cpp_file_path;
   std::vector<float> pcmf32;
   std::vector<std::vector<float>> pcmf32s;
-  whisper_vad_segments *segments;
 
   GetVADContext(self, rwvc);
-  TypedData_Get_Struct(params, ruby_whisper_vad_params, &ruby_whisper_vad_params_type, rwvp);
 
   if (rb_respond_to(file_path, id_to_path)) {
     file_path = rb_funcall(file_path, id_to_path, 0);
@@ -38,12 +36,13 @@ ruby_whisper_vad_detect(VALUE self, VALUE file_path, VALUE params) {
     rb_raise(rb_eRuntimeError, "Failed to open '%s' as WAV file\n", cpp_file_path.c_str());
   }
 
-  segments = whisper_vad_segments_from_samples(rwvc->context, rwvp->params, pcmf32.data(), pcmf32.size());
-  if (segments == nullptr) {
-    rb_raise(rb_eRuntimeError, "Failed to process audio\n");
-  }
-
-  return ruby_whisper_vad_segments_s_init(segments);
+  segments_from_samples_args args = {
+    &self,
+    &params,
+    pcmf32.data(),
+    (int)pcmf32.size(),
+  };
+  return segments_from_samples_body((VALUE)&args);
 }
 
 #ifdef __cplusplus