Merge pull request #225 from m-bain/v3

V3

requirements.txt

@@ -1,10 +1,8 @@
-numpy
-pandas
-torch >=1.9
-torchaudio >=0.10,<1.0
-tqdm
-more-itertools
-transformers>=4.19.0
+torch==2.0.0
+torchaudio==2.0.1
+faster-whisper
+transformers
 ffmpeg-python==0.2.0
-pyannote.audio
-openai-whisper==20230314
+pandas
+setuptools==65.6.3
+nltk

setup.py

@@ -6,8 +6,8 @@
 setup(
     name="whisperx",
     py_modules=["whisperx"],
-    version="2.0.1",
-    description="Time-Accurate Automatic Speech Recognition using Whisper.",
+    version="3.1.0",
+    description="Time-Accurate Automatic Speech Recognition.",
     readme="README.md",
     python_requires=">=3.8",
     author="Max Bain",
@@ -19,7 +19,7 @@
         for r in pkg_resources.parse_requirements(
             open(os.path.join(os.path.dirname(__file__), "requirements.txt"))
         )
-    ],
+    ] + ["pyannote.audio @ git+https://github.com/pyannote/pyannote-audio@11b56a137a578db9335efc00298f6ec1932e6317"],
     entry_points = {
         'console_scripts': ['whisperx=whisperx.transcribe:cli'],
     },

whisperx/__init__.py

@@ -1,3 +1,4 @@
-from .transcribe import transcribe, transcribe_with_vad
+from .transcribe import load_model
 from .alignment import load_align_model, align
-from .vad import load_vad_model
+from .audio import load_audio
+from .diarize import assign_word_speakers, DiarizationPipeline