Skip to content

ruby : Add parallel transcription support #3222

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 15 commits into from
Jun 4, 2025
Merged

ruby : Add parallel transcription support #3222

Show file tree
Hide file tree
Changes from 1 commit
Commits
Show all changes
15 commits
Select commit Hold shift + click to select a range
07a8993
Fix indentation of code sample in document comment
KitaitiMakoto Jun 3, 2025
7cf7ac3
Make Whisper::Context#transcribe able to run non-parallel
KitaitiMakoto Jun 3, 2025
ba9155e
Add test for Whisper::Context#transcribe with parallel option
KitaitiMakoto Jun 3, 2025
c357a81
Follow signature API change of Context#transcribe
KitaitiMakoto Jun 3, 2025
5222a66
Remove useless variable assignment
KitaitiMakoto Jun 3, 2025
7a4ac91
Move simple usage up in README
KitaitiMakoto Jun 3, 2025
b8f966b
Add need help section in README
KitaitiMakoto Jun 3, 2025
628f35e
Add document on Context#transcribe's parallel option in README
KitaitiMakoto Jun 3, 2025
f42317f
Update date
KitaitiMakoto Jun 3, 2025
59aa9d0
Fix signature of Context.new
KitaitiMakoto Jun 3, 2025
81836a5
Make Context#subscribe accept n_processors option
KitaitiMakoto Jun 3, 2025
f95c060
Make test follow #transcribe's change
KitaitiMakoto Jun 3, 2025
19b2757
Make RBS follow #transcribe's change
KitaitiMakoto Jun 3, 2025
a7de86c
Add document for #transcribe's n_processors option
KitaitiMakoto Jun 3, 2025
9f27634
Rename test directory so that Rake tasks' default setting is used
KitaitiMakoto Jun 3, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions bindings/ruby/ext/ruby_whisper.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,7 @@ ID id_URI;
ID id_pre_converted_models;
ID id_coreml_compiled_models;
ID id_cache;
ID id_parallel;

static bool is_log_callback_finalized = false;

Expand Down Expand Up @@ -142,6 +143,7 @@ void Init_whisper() {
id_pre_converted_models = rb_intern("pre_converted_models");
id_coreml_compiled_models = rb_intern("coreml_compiled_models");
id_cache = rb_intern("cache");
id_parallel = rb_intern("parallel");

mWhisper = rb_define_module("Whisper");
mVAD = rb_define_module_under(mWhisper, "VAD");
Expand Down
5 changes: 5 additions & 0 deletions bindings/ruby/ext/ruby_whisper_context.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,7 @@ extern ID id_URI;
extern ID id_pre_converted_models;
extern ID id_coreml_compiled_models;
extern ID id_cache;
extern ID id_parallel;

extern VALUE cContext;
extern VALUE eError;
Expand All @@ -24,6 +25,8 @@ extern VALUE rb_whisper_model_s_new(VALUE context);
extern VALUE rb_whisper_segment_s_new(VALUE context, int index);
extern void prepare_transcription(ruby_whisper_params *rwp, VALUE *context);

ID transcribe_option_names[1];

static void
ruby_whisper_free(ruby_whisper *rw)
{
Expand Down Expand Up @@ -633,6 +636,8 @@ init_ruby_whisper_context(VALUE *mWhisper)
{
cContext = rb_define_class_under(*mWhisper, "Context", rb_cObject);

transcribe_option_names[0] = id_parallel;

rb_define_alloc_func(cContext, ruby_whisper_allocate);
rb_define_method(cContext, "initialize", ruby_whisper_initialize, -1);

Expand Down
19 changes: 15 additions & 4 deletions bindings/ruby/ext/ruby_whisper_transcribe.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,6 +13,7 @@ extern const rb_data_type_t ruby_whisper_params_type;

extern ID id_to_s;
extern ID id_call;
extern ID transcribe_option_names[1];

extern void
prepare_transcription(ruby_whisper_params * rwp, VALUE * self);
Expand All @@ -34,9 +35,14 @@ VALUE
ruby_whisper_transcribe(int argc, VALUE *argv, VALUE self) {
ruby_whisper *rw;
ruby_whisper_params *rwp;
VALUE wave_file_path, blk, params;
VALUE wave_file_path, blk, params, kws;
VALUE opts[1];

rb_scan_args_kw(RB_SCAN_ARGS_LAST_HASH_KEYWORDS, argc, argv, "2:&", &wave_file_path, &params, &kws, &blk);
rb_get_kwargs(kws, transcribe_option_names, 0, 1, opts);

bool parallel = !(NIL_P(opts[0]) || opts[0] == Qfalse);

rb_scan_args(argc, argv, "02&", &wave_file_path, &params, &blk);
TypedData_Get_Struct(self, ruby_whisper, &ruby_whisper_type, rw);
TypedData_Get_Struct(params, ruby_whisper_params, &ruby_whisper_params_type, rwp);

Expand Down Expand Up @@ -66,8 +72,13 @@ ruby_whisper_transcribe(int argc, VALUE *argv, VALUE self) {

prepare_transcription(rwp, &self);

if (whisper_full_parallel(rw->context, rwp->params, pcmf32.data(), pcmf32.size(), 1) != 0) {
fprintf(stderr, "failed to process audio\n");
int result;
if (parallel) {
result = whisper_full_parallel(rw->context, rwp->params, pcmf32.data(), pcmf32.size(), 1);
} else {
result = whisper_full(rw->context, rwp->params, pcmf32.data(), pcmf32.size());
}
if (result != 0) {
return self;
}
const int n_segments = whisper_full_n_segments(rw->context);
Expand Down