ailia-speech 1.4.0__tar.gz → 1.4.0.1__tar.gz

{ailia_speech-1.4.0 → ailia_speech-1.4.0.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ailia_speech
-Version: 1.4.0
+Version: 1.4.0.1
 Summary: ailia AI Speech
 Home-page: https://ailia.jp/
 Author: ax Inc.
@@ -117,7 +117,7 @@ for i in range(0, audio_waveform.shape[0], sampling_rate):
 By specifying dialization_type, speaker diarization can be performed. When speaker diarization is enabled, speaker_id becomes valid.
 
 ```
-speech.initialize_model(model_path = "./models/", model_type = ailia_speech.AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_LARGE_V3_TURBO, dialization_type = AILIA_SPEECH_DIARIZATION_TYPE_PYANNOTE_AUDIO)
+speech.initialize_model(model_path = "./models/", model_type = ailia_speech.AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_LARGE_V3_TURBO, diarization_type = ailia_speech.AILIA_SPEECH_DIARIZATION_TYPE_PYANNOTE_AUDIO)
 ```
 
 ### Available model types

{ailia_speech-1.4.0 → ailia_speech-1.4.0.1}/README.md

@@ -95,7 +95,7 @@ for i in range(0, audio_waveform.shape[0], sampling_rate):
 By specifying dialization_type, speaker diarization can be performed. When speaker diarization is enabled, speaker_id becomes valid.
 
 ```
-speech.initialize_model(model_path = "./models/", model_type = ailia_speech.AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_LARGE_V3_TURBO, dialization_type = AILIA_SPEECH_DIARIZATION_TYPE_PYANNOTE_AUDIO)
+speech.initialize_model(model_path = "./models/", model_type = ailia_speech.AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_LARGE_V3_TURBO, diarization_type = ailia_speech.AILIA_SPEECH_DIARIZATION_TYPE_PYANNOTE_AUDIO)
 ```
 
 ### Available model types

{ailia_speech-1.4.0 → ailia_speech-1.4.0.1}/ailia_speech/__init__.py

@@ -409,7 +409,7 @@ class Whisper(AiliaSpeechModel):
             intermediate_callback_cnt = intermediate_callback_cnt + 1
 
 
-    def initialize_model(self, model_path = "./", model_type = AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_TINY, vad_type = AILIA_SPEECH_VAD_TYPE_SILERO, dialization_type = None):
+    def initialize_model(self, model_path = "./", model_type = AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_TINY, vad_type = AILIA_SPEECH_VAD_TYPE_SILERO, diarization_type = None):
         if "time_license" in ailia.get_version():
             ailia.check_and_download_license()
         if model_type == AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_TINY:
@@ -448,15 +448,15 @@ class Whisper(AiliaSpeechModel):
             encoder_pb_path = "encoder_turbo_weights.opt.pb"
             decoder_pb_path = None
             model_type = AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_LARGE_V3
-        self._download_model(model_path, encoder_path, decoder_path, encoder_pb_path, decoder_pb_path, vad_type, dialization_type)
+        self._download_model(model_path, encoder_path, decoder_path, encoder_pb_path, decoder_pb_path, vad_type, diarization_type)
         self._open_model(model_path + encoder_path, model_path + decoder_path, model_type)
         if vad_type is not None:
             self._open_vad(model_path + "silero_vad.onnx", vad_type)
-        if dialization_type is not None:
-            self._open_diarization(model_path + "segmentation.onnx", model_path + "speaker-embedding.onnx", dialization_type)
+        if diarization_type is not None:
+            self._open_diarization(model_path + "segmentation.onnx", model_path + "speaker-embedding.onnx", diarization_type)
 
 
-    def _download_model(self, model_path, encoder_path, decoder_path, encoder_pb_path, decoder_pb_path, vad_type, dialization_type):
+    def _download_model(self, model_path, encoder_path, decoder_path, encoder_pb_path, decoder_pb_path, vad_type, diarization_type):
         REMOTE_PATH = "https://storage.googleapis.com/ailia-models/whisper/"
         os.makedirs(model_path, exist_ok = True)
         check_and_download_file(model_path + encoder_path, REMOTE_PATH)
@@ -470,7 +470,7 @@ class Whisper(AiliaSpeechModel):
             REMOTE_PATH = "https://storage.googleapis.com/ailia-models/silero-vad/"
             check_and_download_file(model_path + "silero_vad.onnx", REMOTE_PATH)
 
-        if dialization_type is not None:
+        if diarization_type is not None:
             REMOTE_PATH = "https://storage.googleapis.com/ailia-models/pyannote-audio/"
             check_and_download_file(model_path + "segmentation.onnx", REMOTE_PATH)
             check_and_download_file(model_path + "speaker-embedding.onnx", REMOTE_PATH)

{ailia_speech-1.4.0 → ailia_speech-1.4.0.1}/ailia_speech.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ailia_speech
-Version: 1.4.0
+Version: 1.4.0.1
 Summary: ailia AI Speech
 Home-page: https://ailia.jp/
 Author: ax Inc.
@@ -117,7 +117,7 @@ for i in range(0, audio_waveform.shape[0], sampling_rate):
 By specifying dialization_type, speaker diarization can be performed. When speaker diarization is enabled, speaker_id becomes valid.
 
 ```
-speech.initialize_model(model_path = "./models/", model_type = ailia_speech.AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_LARGE_V3_TURBO, dialization_type = AILIA_SPEECH_DIARIZATION_TYPE_PYANNOTE_AUDIO)
+speech.initialize_model(model_path = "./models/", model_type = ailia_speech.AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_LARGE_V3_TURBO, diarization_type = ailia_speech.AILIA_SPEECH_DIARIZATION_TYPE_PYANNOTE_AUDIO)
 ```
 
 ### Available model types

{ailia_speech-1.4.0 → ailia_speech-1.4.0.1}/setup.py

@@ -54,7 +54,7 @@ if __name__ == "__main__":
     setup(
         name="ailia_speech",
         scripts=scripts,
-        version="1.4.0",
+        version="1.4.0.1",
         install_requires=[
             "ailia",
             "ailia_tokenizer",