ailia-speech 1.3.1.0__py3-none-any.whl → 1.3.2.0__py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Potentially problematic release.


This version of ailia-speech might be problematic. Click here for more details.

ailia_speech/__init__.py CHANGED
@@ -261,6 +261,9 @@ dll.ailiaSpeechSetIntermediateCallback.argtypes = (c_void_p, AILIA_SPEECH_USER_A
261
261
  dll.ailiaSpeechSetLanguage.restype = c_int
262
262
  dll.ailiaSpeechSetLanguage.argtypes = (c_void_p, c_char_p)
263
263
 
264
+ dll.ailiaSpeechSetSilentThreshold.restype = c_int
265
+ dll.ailiaSpeechSetSilentThreshold.argtypes = (c_void_p, c_float, c_float, c_float)
266
+
264
267
  # ==============================================================================
265
268
  # model download
266
269
  # ==============================================================================
@@ -466,6 +469,9 @@ class Whisper(AiliaSpeechModel):
466
469
  else:
467
470
  self._check(dll.ailiaSpeechOpenVadFileA(self._instance, p1, vad_type))
468
471
 
472
+ def set_silent_threshold(self, silent_threshold, speech_sec, no_speech_sec):
473
+ self._check(dll.ailiaSpeechSetSilentThreshold(self._instance, silent_threshold, speech_sec, no_speech_sec))
474
+
469
475
  def transcribe(self, audio_waveform, sampling_rate, lang = None):
470
476
  if len(audio_waveform.shape) == 1:
471
477
  channels = 1
Binary file
Binary file
Binary file
@@ -261,6 +261,9 @@ dll.ailiaSpeechSetIntermediateCallback.argtypes = (c_void_p, AILIA_SPEECH_USER_A
261
261
  dll.ailiaSpeechSetLanguage.restype = c_int
262
262
  dll.ailiaSpeechSetLanguage.argtypes = (c_void_p, c_char_p)
263
263
 
264
+ dll.ailiaSpeechSetSilentThreshold.restype = c_int
265
+ dll.ailiaSpeechSetSilentThreshold.argtypes = (c_void_p, c_float, c_float, c_float)
266
+
264
267
  # ==============================================================================
265
268
  # model download
266
269
  # ==============================================================================
@@ -466,6 +469,9 @@ class Whisper(AiliaSpeechModel):
466
469
  else:
467
470
  self._check(dll.ailiaSpeechOpenVadFileA(self._instance, p1, vad_type))
468
471
 
472
+ def set_silent_threshold(self, silent_threshold, speech_sec, no_speech_sec):
473
+ self._check(dll.ailiaSpeechSetSilentThreshold(self._instance, silent_threshold, speech_sec, no_speech_sec))
474
+
469
475
  def transcribe(self, audio_waveform, sampling_rate, lang = None):
470
476
  if len(audio_waveform.shape) == 1:
471
477
  channels = 1
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.1
2
2
  Name: ailia_speech
3
- Version: 1.3.1.0
3
+ Version: 1.3.2.0
4
4
  Summary: ailia AI Speech
5
5
  Home-page: https://ailia.jp/
6
6
  Author: ax Inc.
@@ -93,6 +93,7 @@ audio_waveform, sampling_rate = librosa.load(input_file_path, mono = True)
93
93
  # Infer
94
94
  speech = ailia_speech.Whisper()
95
95
  speech.initialize_model(model_path = "./models/", model_type = ailia_speech.AILIA_SPEECH_MODEL_TYPE_WHISPER_MULTILINGUAL_LARGE_V3_TURBO)
96
+ speech.set_silent_threshold(silent_threshold = 0.5, speech_sec = 1.0, no_speech_sec = 0.5)
96
97
  for i in range(0, audio_waveform.shape[0], sampling_rate):
97
98
  complete = False
98
99
  if i + sampling_rate >= audio_waveform.shape[0]:
@@ -0,0 +1,12 @@
1
+ ailia_speech/LICENSE_AILIA_EN.pdf,sha256=1DzVViPnw1uAS8gJ5a8uN3iZNNR5I1ItIXmezHfUpeM,70149
2
+ ailia_speech/LICENSE_AILIA_JA.pdf,sha256=s628QN47S2bNqIfuSjm2LBf0vIluv2df6MSemn6Ksmw,174134
3
+ ailia_speech/__init__.py,sha256=xPkxrpftgloJIsJtnAYvrn7TYemZDmnQRQFBY2sw_AM,28247
4
+ ailia_speech/linux/arm64-v8a/libailia_speech.so,sha256=6EMZQdC3EUvdwJnNbAfcZ1FZpQuKu68e9M1tmFXzPTo,170296
5
+ ailia_speech/linux/x64/libailia_speech.so,sha256=1KiajIRrHK4J5UMEi8MWXP-rfofyEhsA-qWk1f3QSOk,171360
6
+ ailia_speech/mac/libailia_speech.dylib,sha256=sH5JXhBUpIXDeZtuZne_It7J0TAQiOfNhJhyxDvuHrc,317112
7
+ ailia_speech/windows/x64/ailia_speech.dll,sha256=CNVJWs7yI-TvbVCOuS5V9rIRPh2GlEBsNrSDst1aF0w,126976
8
+ ailia_speech-1.3.2.0.data/scripts/__init__.py,sha256=xPkxrpftgloJIsJtnAYvrn7TYemZDmnQRQFBY2sw_AM,28247
9
+ ailia_speech-1.3.2.0.dist-info/METADATA,sha256=SUUpnJOp4kXWLpRZLZqkj7U9vfXRLEowKYnB9iVh9LU,3701
10
+ ailia_speech-1.3.2.0.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
11
+ ailia_speech-1.3.2.0.dist-info/top_level.txt,sha256=Ou9XeJ9AvdK8eutw07oosCthftD1tRYzAgNY2BrYhDc,13
12
+ ailia_speech-1.3.2.0.dist-info/RECORD,,
@@ -1,12 +0,0 @@
1
- ailia_speech/LICENSE_AILIA_EN.pdf,sha256=1DzVViPnw1uAS8gJ5a8uN3iZNNR5I1ItIXmezHfUpeM,70149
2
- ailia_speech/LICENSE_AILIA_JA.pdf,sha256=s628QN47S2bNqIfuSjm2LBf0vIluv2df6MSemn6Ksmw,174134
3
- ailia_speech/__init__.py,sha256=NMFB_mXj9ed9YcbEXrjvm-82OOYRxzSBENKCuxbtz6M,27915
4
- ailia_speech/linux/arm64-v8a/libailia_speech.so,sha256=JAOwnBr7lbiMZmPCM99pd4vJQ08ZuXDPpq-FurrXSnE,166096
5
- ailia_speech/linux/x64/libailia_speech.so,sha256=WbFvA5wKTgS_Zx8ErT7WBKJbzOUexavr4nP4EkLNawQ,171360
6
- ailia_speech/mac/libailia_speech.dylib,sha256=-JAC40yLslAVMvfh6LhDvP3Zyt3hIT3WZc7wa9-07zU,317112
7
- ailia_speech/windows/x64/ailia_speech.dll,sha256=WJCOHi0Na4tdMG1RT7dA7yAoWumiGSWeW1vxUtiXDS8,126464
8
- ailia_speech-1.3.1.0.data/scripts/__init__.py,sha256=NMFB_mXj9ed9YcbEXrjvm-82OOYRxzSBENKCuxbtz6M,27915
9
- ailia_speech-1.3.1.0.dist-info/METADATA,sha256=x5WBVorX7b9Mubn69FVh75XS2b3iDIy7ZM38F9yyGRo,3610
10
- ailia_speech-1.3.1.0.dist-info/WHEEL,sha256=GV9aMThwP_4oNCtvEC2ec3qUYutgWeAzklro_0m4WJQ,91
11
- ailia_speech-1.3.1.0.dist-info/top_level.txt,sha256=Ou9XeJ9AvdK8eutw07oosCthftD1tRYzAgNY2BrYhDc,13
12
- ailia_speech-1.3.1.0.dist-info/RECORD,,