Merge pull request #85 from tazz4843/whisper-8e46ba8

Update `whisper.cpp` to 8e46ba8

README.md

@@ -67,7 +67,7 @@ See [BUILDING.md](BUILDING.md) for instructions for building whisper-rs on Windo
 ## Troubleshooting
 
 * Something other than Windows/macOS/Linux isn't working!
-  * I don't have a way to test this platforms, so I can't really help you.
+  * I don't have a way to test these platforms, so I can't really help you.
     * If you can get it working, please open a PR with any changes to make it work and build instructions in BUILDING.md!
 * I get a panic during binding generation build!
   * You can attempt to fix it yourself, or you can set the `WHISPER_DONT_GENERATE_BINDINGS` environment variable.

src/lib.rs

@@ -24,3 +24,5 @@ pub type WhisperNewSegmentCallback = whisper_rs_sys::whisper_new_segment_callbac
 pub type WhisperStartEncoderCallback = whisper_rs_sys::whisper_encoder_begin_callback;
 pub type WhisperProgressCallback = whisper_rs_sys::whisper_progress_callback;
 pub type WhisperLogitsFilterCallback = whisper_rs_sys::whisper_logits_filter_callback;
+pub type WhisperAbortCallback = whisper_rs_sys::whisper_abort_callback;
+pub type WhisperLogCallback = whisper_rs_sys::whisper_log_callback;

src/standalone.rs

@@ -1,6 +1,5 @@
 //! Standalone functions that have no associated type.
 
-use crate::WhisperToken;
 use std::ffi::{c_int, CStr, CString};
 
 /// Return the id of the specified language, returns -1 if not found
@@ -54,21 +53,12 @@ pub fn get_lang_str(id: i32) -> Option<&'static str> {
     }
 }
 
-// task tokens
-/// Get the ID of the translate task token.
+/// Callback to control logging output: default behaviour is to print to stderr.
 ///
 /// # C++ equivalent
-/// `whisper_token whisper_token_translate ()`
-pub fn token_translate() -> WhisperToken {
-    unsafe { whisper_rs_sys::whisper_token_translate() }
-}
-
-/// Get the ID of the transcribe task token.
-///
-/// # C++ equivalent
-/// `whisper_token whisper_token_transcribe()`
-pub fn token_transcribe() -> WhisperToken {
-    unsafe { whisper_rs_sys::whisper_token_transcribe() }
+/// `void whisper_set_log_callback(whisper_log_callback callback);`
+pub unsafe fn set_log_callback(callback: whisper_rs_sys::whisper_log_callback) {
+    unsafe { whisper_rs_sys::whisper_set_log_callback(callback) }
 }
 
 /// Print system information.

src/whisper_ctx.rs

@@ -367,6 +367,15 @@ impl WhisperContext {
         unsafe { whisper_rs_sys::whisper_token_sot(self.ctx) }
     }
 
+    /// Get the ID of the solm token.
+    ///
+    /// # C++ equivalent
+    /// `whisper_token whisper_token_solm(struct whisper_context * ctx)`
+    #[inline]
+    pub fn token_solm(&self) -> WhisperToken {
+        unsafe { whisper_rs_sys::whisper_token_solm(self.ctx) }
+    }
+
     /// Get the ID of the prev token.
     ///
     /// # C++ equivalent
@@ -376,13 +385,13 @@ impl WhisperContext {
         unsafe { whisper_rs_sys::whisper_token_prev(self.ctx) }
     }
 
-    /// Get the ID of the solm token.
+    /// Get the ID of the nosp token.
     ///
     /// # C++ equivalent
-    /// `whisper_token whisper_token_solm(struct whisper_context * ctx)`
+    /// `whisper_token whisper_token_nosp(struct whisper_context * ctx)`
     #[inline]
-    pub fn token_solm(&self) -> WhisperToken {
-        unsafe { whisper_rs_sys::whisper_token_solm(self.ctx) }
+    pub fn token_nosp(&self) -> WhisperToken {
+        unsafe { whisper_rs_sys::whisper_token_nosp(self.ctx) }
     }
 
     /// Get the ID of the not token.
@@ -432,6 +441,23 @@ impl WhisperContext {
     pub fn reset_timings(&self) {
         unsafe { whisper_rs_sys::whisper_reset_timings(self.ctx) }
     }
+
+    // task tokens
+    /// Get the ID of the translate task token.
+    ///
+    /// # C++ equivalent
+    /// `whisper_token whisper_token_translate ()`
+    pub fn token_translate(&self) -> WhisperToken {
+        unsafe { whisper_rs_sys::whisper_token_translate(self.ctx) }
+    }
+
+    /// Get the ID of the transcribe task token.
+    ///
+    /// # C++ equivalent
+    /// `whisper_token whisper_token_transcribe()`
+    pub fn token_transcribe(&self) -> WhisperToken {
+        unsafe { whisper_rs_sys::whisper_token_transcribe(self.ctx) }
+    }
 }
 
 impl Drop for WhisperContext {

src/whisper_params.rs

@@ -206,6 +206,15 @@ impl<'a, 'b> FullParams<'a, 'b> {
         self.fp.speed_up = speed_up;
     }
 
+    /// # EXPERIMENTAL
+    ///
+    /// Enables debug mode, such as dumping the log mel spectrogram.
+    ///
+    /// Defaults to false.
+    pub fn set_debug_mode(&mut self, debug: bool) {
+        self.fp.debug_mode = debug;
+    }
+
     /// # EXPERIMENTAL
     ///
     /// Overwrite the audio context size. 0 = default.
@@ -216,6 +225,16 @@ impl<'a, 'b> FullParams<'a, 'b> {
         self.fp.audio_ctx = audio_ctx;
     }
 
+    /// # EXPERIMENTAL
+    ///
+    /// Enable tinydiarize support.
+    /// Experimental speaker turn detection.
+    ///
+    /// Defaults to false.
+    pub fn set_tdrz_enable(&mut self, tdrz_enable: bool) {
+        self.fp.tdrz_enable = tdrz_enable;
+    }
+
     /// Set tokens to provide the model as initial input.
     ///
     /// These tokens are prepended to any existing text content from a previous call.
@@ -485,6 +504,31 @@ impl<'a, 'b> FullParams<'a, 'b> {
     ) {
         self.fp.logits_filter_callback_user_data = user_data;
     }
+
+    /// Set the callback that is called each time before ggml computation starts.
+    ///
+    /// Note that this callback has not been Rustified yet (and likely never will be, unless someone else feels the need to do so).
+    /// It is still a C callback.
+    ///
+    /// # Safety
+    /// Do not use this function unless you know what you are doing.
+    /// * Be careful not to mutate the state of the whisper_context pointer returned in the callback.
+    ///   This could cause undefined behavior, as this violates the thread-safety guarantees of the underlying C library.
+    ///
+    /// Defaults to None.
+    pub unsafe fn set_abort_callback(&mut self, abort_callback: crate::WhisperAbortCallback) {
+        self.fp.abort_callback = abort_callback;
+    }
+
+    /// Set the user data to be passed to the abort callback.
+    ///
+    /// # Safety
+    /// See the safety notes for `set_abort_callback`.
+    ///
+    /// Defaults to None.
+    pub unsafe fn set_abort_callback_user_data(&mut self, user_data: *mut std::ffi::c_void) {
+        self.fp.abort_callback_user_data = user_data;
+    }
 }
 
 // following implementations are safe