isort

whisperlivekit/__init__.py

@@ -1,7 +1,7 @@
 from .audio_processor import AudioProcessor
 from .core import TranscriptionEngine
 from .parse_args import parse_args
-from .web.web_interface import get_web_interface_html, get_inline_ui_html
+from .web.web_interface import get_inline_ui_html, get_web_interface_html
 
 __all__ = [
     "TranscriptionEngine",

whisperlivekit/audio_processor.py

@@ -1,14 +1,20 @@
 import asyncio
-import numpy as np
-from time import time
 import logging
 import traceback
-from typing import Optional, Union, List, Any, AsyncGenerator
+from time import time
-from whisperlivekit.timed_objects import ASRToken, Silence, Line, FrontData, State, Transcript, ChangeSpeaker
+from typing import Any, AsyncGenerator, List, Optional, Union
-from whisperlivekit.core import TranscriptionEngine, online_factory, online_diarization_factory, online_translation_factory
+
-from whisperlivekit.silero_vad_iterator import FixedVADIterator
+import numpy as np
+
+from whisperlivekit.core import (TranscriptionEngine,
+                                 online_diarization_factory, online_factory,
+                                 online_translation_factory)
 from whisperlivekit.ffmpeg_manager import FFmpegManager, FFmpegState
+from whisperlivekit.silero_vad_iterator import FixedVADIterator
+from whisperlivekit.timed_objects import (ASRToken, ChangeSpeaker, FrontData,
+                                          Line, Silence, State, Transcript)
 from whisperlivekit.tokens_alignment import TokensAlignment
+
 logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s")
 logger = logging.getLogger(__name__)
 logger.setLevel(logging.DEBUG)

whisperlivekit/basic_server.py

@@ -1,10 +1,13 @@
-from contextlib import asynccontextmanager
-from fastapi import FastAPI, WebSocket, WebSocketDisconnect
-from fastapi.responses import HTMLResponse
-from fastapi.middleware.cors import CORSMiddleware
-from whisperlivekit import TranscriptionEngine, AudioProcessor, get_inline_ui_html, parse_args
 import asyncio
 import logging
+from contextlib import asynccontextmanager
+
+from fastapi import FastAPI, WebSocket, WebSocketDisconnect
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.responses import HTMLResponse
+
+from whisperlivekit import (AudioProcessor, TranscriptionEngine,
+                            get_inline_ui_html, parse_args)
 
 logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s")
 logging.getLogger().setLevel(logging.WARNING)

whisperlivekit/core.py

@@ -1,9 +1,11 @@
+import logging
+import sys
+from argparse import Namespace
+
+from whisperlivekit.local_agreement.online_asr import OnlineASRProcessor
 from whisperlivekit.local_agreement.whisper_online import backend_factory
 from whisperlivekit.simul_whisper import SimulStreamingASR
-from whisperlivekit.local_agreement.online_asr import OnlineASRProcessor
+
-from argparse import Namespace
-import sys
-import logging
 
 def update_with_kwargs(_dict, kwargs):
     _dict.update({
@@ -80,6 +82,7 @@ class TranscriptionEngine:
         
         if self.args.vac:
             from whisperlivekit.silero_vad_iterator import load_silero_vad
+
             # Use ONNX if specified, otherwise use JIT (default)
             use_onnx = kwargs.get('vac_onnx', False)
             self.vac_model = load_silero_vad(onnx=use_onnx)
@@ -135,7 +138,8 @@ class TranscriptionEngine:
 
         if self.args.diarization:
             if self.args.diarization_backend == "diart":
-                from whisperlivekit.diarization.diart_backend import DiartDiarization
+                from whisperlivekit.diarization.diart_backend import \
+                    DiartDiarization
                 diart_params = {
                     "segmentation_model": "pyannote/segmentation-3.0",
                     "embedding_model": "pyannote/embedding",
@@ -146,7 +150,8 @@ class TranscriptionEngine:
                     **diart_params
                 )
             elif self.args.diarization_backend == "sortformer":
-                from whisperlivekit.diarization.sortformer_backend import SortformerDiarization
+                from whisperlivekit.diarization.sortformer_backend import \
+                    SortformerDiarization
                 self.diarization_model = SortformerDiarization()
         
         self.translation_model = None
@@ -182,7 +187,8 @@ def online_diarization_factory(args, diarization_backend):
         # Not the best here, since several user/instances will share the same backend, but diart is not SOTA anymore and sortformer is recommended
     
     if args.diarization_backend == "sortformer":
-        from whisperlivekit.diarization.sortformer_backend import SortformerDiarizationOnline
+        from whisperlivekit.diarization.sortformer_backend import \
+            SortformerDiarizationOnline
         online = SortformerDiarizationOnline(shared_model=diarization_backend)
     return online
 

whisperlivekit/diarization/diart_backend.py

@@ -1,20 +1,20 @@
 import asyncio
+import logging
 import re
 import threading
-import numpy as np
-import logging
 import time
-from queue import SimpleQueue, Empty
+from queue import Empty, SimpleQueue
+from typing import Any, List, Tuple
 
+import diart.models as m
+import numpy as np
 from diart import SpeakerDiarization, SpeakerDiarizationConfig
 from diart.inference import StreamingInference
-from diart.sources import AudioSource
+from diart.sources import AudioSource, MicrophoneAudioSource
-from whisperlivekit.timed_objects import SpeakerSegment
-from diart.sources import MicrophoneAudioSource
-from rx.core import Observer
-from typing import Tuple, Any, List
 from pyannote.core import Annotation
-import diart.models as m
+from rx.core import Observer
+
+from whisperlivekit.timed_objects import SpeakerSegment
 
 logger = logging.getLogger(__name__)
 

whisperlivekit/diarization/sortformer_backend.py

@@ -1,11 +1,12 @@
-import numpy as np
-import torch
 import logging
 import threading
 import time
 import wave
+from queue import Empty, SimpleQueue
 from typing import List, Optional
-from queue import SimpleQueue, Empty
+
+import numpy as np
+import torch
 
 from whisperlivekit.timed_objects import SpeakerSegment
 
@@ -295,6 +296,7 @@ def extract_number(s: str) -> int:
 
 if __name__ == '__main__':
     import asyncio
+
     import librosa
     
     async def main():

whisperlivekit/ffmpeg_manager.py

@@ -1,8 +1,8 @@
 import asyncio
+import contextlib
 import logging
 from enum import Enum
-from typing import Optional, Callable
+from typing import Callable, Optional
-import contextlib
 
 logger = logging.getLogger(__name__)
 logging.basicConfig(level=logging.INFO)

whisperlivekit/local_agreement/backends.py

@@ -1,13 +1,16 @@
-import sys
-import logging
 import io
-import soundfile as sf
+import logging
 import math
+import sys
 from typing import List
+
 import numpy as np
+import soundfile as sf
+
+from whisperlivekit.model_paths import model_path_and_type, resolve_model_path
 from whisperlivekit.timed_objects import ASRToken
-from whisperlivekit.model_paths import resolve_model_path, model_path_and_type
 from whisperlivekit.whisper.transcribe import transcribe as whisper_transcribe
+
 logger = logging.getLogger(__name__)
 class ASRBase:
     sep = " "  # join transcribe words with this character (" " for whisper_timestamped,
@@ -165,8 +168,8 @@ class MLXWhisper(ASRBase):
     sep = ""
 
     def load_model(self, model_size=None, cache_dir=None, model_dir=None):
-        from mlx_whisper.transcribe import ModelHolder, transcribe
         import mlx.core as mx
+        from mlx_whisper.transcribe import ModelHolder, transcribe
 
         if model_dir is not None:
             resolved_path = resolve_model_path(model_dir)

whisperlivekit/local_agreement/online_asr.py

@@ -1,7 +1,9 @@
-import sys
-import numpy as np
 import logging
-from typing import List, Tuple, Optional
+import sys
+from typing import List, Optional, Tuple
+
+import numpy as np
+
 from whisperlivekit.timed_objects import ASRToken, Sentence, Transcript
 
 logger = logging.getLogger(__name__)

whisperlivekit/local_agreement/whisper_online.py

@@ -1,18 +1,19 @@
 #!/usr/bin/env python3
-import sys
-import numpy as np
-import librosa
-from functools import lru_cache
-import time
 import logging
 import platform
-from .backends import FasterWhisperASR, MLXWhisper, WhisperASR, OpenaiApiASR
+import sys
+import time
+from functools import lru_cache
+
+import librosa
+import numpy as np
+
+from whisperlivekit.backend_support import (faster_backend_available,
+                                            mlx_backend_available)
+from whisperlivekit.model_paths import model_path_and_type, resolve_model_path
 from whisperlivekit.warmup import warmup_asr
-from whisperlivekit.model_paths import resolve_model_path, model_path_and_type
+
-from whisperlivekit.backend_support import (
+from .backends import FasterWhisperASR, MLXWhisper, OpenaiApiASR, WhisperASR
-    mlx_backend_available,
-    faster_backend_available,
-)
 
 logger = logging.getLogger(__name__)
 

whisperlivekit/parse_args.py

@@ -1,6 +1,7 @@
 
 from argparse import ArgumentParser
 
+
 def parse_args():
     parser = ArgumentParser(description="Whisper FastAPI Online Server")
     parser.add_argument(

whisperlivekit/silero_vad_iterator.py

@@ -1,8 +1,9 @@
-import torch
-import numpy as np
 import warnings
 from pathlib import Path
 
+import numpy as np
+import torch
+
 """
 Code is adapted from silero-vad v6: https://github.com/snakers4/silero-vad
 """

whisperlivekit/simul_whisper/backend.py

@@ -1,31 +1,30 @@
-import sys
+import gc
-import numpy as np
 import logging
-from typing import List, Tuple, Optional
+import os
 import platform
-from whisperlivekit.timed_objects import ASRToken, Transcript, ChangeSpeaker
+import sys
+from pathlib import Path
+from typing import List, Optional, Tuple
+
+import numpy as np
+import torch
+
+from whisperlivekit.backend_support import (faster_backend_available,
+                                            mlx_backend_available)
+from whisperlivekit.model_paths import model_path_and_type, resolve_model_path
+from whisperlivekit.simul_whisper.config import AlignAttConfig
+from whisperlivekit.simul_whisper.simul_whisper import AlignAtt
+from whisperlivekit.timed_objects import ASRToken, ChangeSpeaker, Transcript
 from whisperlivekit.warmup import load_file
 from whisperlivekit.whisper import load_model, tokenizer
 from whisperlivekit.whisper.audio import TOKENS_PER_SECOND
-import os
-import gc
-from pathlib import Path
-from whisperlivekit.model_paths import model_path_and_type, resolve_model_path
-from whisperlivekit.backend_support import (
-    mlx_backend_available,
-    faster_backend_available,
-)
-
-import torch
-from whisperlivekit.simul_whisper.config import AlignAttConfig
-from whisperlivekit.simul_whisper.simul_whisper import AlignAtt
 
 logger = logging.getLogger(__name__)
 
 
 HAS_MLX_WHISPER = mlx_backend_available(warn_on_missing=True)
 if HAS_MLX_WHISPER:
-    from .mlx_encoder import mlx_model_mapping, load_mlx_encoder
+    from .mlx_encoder import load_mlx_encoder, mlx_model_mapping
 else:
     mlx_model_mapping = {}
 HAS_FASTER_WHISPER = faster_backend_available(warn_on_missing=not HAS_MLX_WHISPER)

whisperlivekit/simul_whisper/beam.py

@@ -1,5 +1,6 @@
 from whisperlivekit.whisper.decoding import PyTorchInference
 
+
 # extention of PyTorchInference for beam search
 class BeamPyTorchInference(PyTorchInference):
 

whisperlivekit/simul_whisper/config.py

@@ -1,6 +1,7 @@
 from dataclasses import dataclass, field
 from typing import Literal
 
+
 @dataclass
 class AlignAttConfig():
     eval_data_path: str = "tmp"

whisperlivekit/simul_whisper/mlx_encoder.py

@@ -5,7 +5,6 @@ import mlx.core as mx
 import mlx.nn as nn
 from huggingface_hub import snapshot_download
 from mlx.utils import tree_unflatten
-
 from mlx_whisper import whisper
 
 mlx_model_mapping = {

whisperlivekit/simul_whisper/simul_whisper.py

@@ -1,33 +1,34 @@
-import os
 import logging
-
-import torch
-import torch.nn.functional as F
-import numpy as np
-
-from whisperlivekit.whisper import DecodingOptions, tokenizer
-from .config import AlignAttConfig
-from whisperlivekit.timed_objects import ASRToken
-from whisperlivekit.whisper.audio import log_mel_spectrogram, TOKENS_PER_SECOND, pad_or_trim, N_SAMPLES, N_FRAMES
-from whisperlivekit.whisper.timing import median_filter
-from whisperlivekit.whisper.decoding import GreedyDecoder, BeamSearchDecoder, SuppressTokens
-from .beam import BeamPyTorchInference
-from .eow_detection import fire_at_boundary, load_cif
 import os
 from time import time
-from .token_buffer import TokenBuffer
+
-from whisperlivekit.backend_support import (
+import numpy as np
-    mlx_backend_available,
+import torch
-    faster_backend_available,
+import torch.nn.functional as F
-)
+
+from whisperlivekit.backend_support import (faster_backend_available,
+                                            mlx_backend_available)
+from whisperlivekit.timed_objects import ASRToken
+from whisperlivekit.whisper import DecodingOptions, tokenizer
+from whisperlivekit.whisper.audio import (N_FRAMES, N_SAMPLES,
+                                          TOKENS_PER_SECOND,
+                                          log_mel_spectrogram, pad_or_trim)
+from whisperlivekit.whisper.decoding import (BeamSearchDecoder, GreedyDecoder,
+                                             SuppressTokens)
+from whisperlivekit.whisper.timing import median_filter
 
 from ..timed_objects import PUNCTUATION_MARKS
+from .beam import BeamPyTorchInference
+from .config import AlignAttConfig
+from .eow_detection import fire_at_boundary, load_cif
+from .token_buffer import TokenBuffer
 
 DEC_PAD = 50257
 logger = logging.getLogger(__name__)
 
 if mlx_backend_available():
-    from mlx_whisper.audio import log_mel_spectrogram as mlx_log_mel_spectrogram
+    from mlx_whisper.audio import \
+        log_mel_spectrogram as mlx_log_mel_spectrogram
     from mlx_whisper.transcribe import pad_or_trim as mlx_pad_or_trim
 
 if faster_backend_available():

whisperlivekit/simul_whisper/token_buffer.py

@@ -1,5 +1,8 @@
-import torch
 import sys
+
+import torch
+
+
 class TokenBuffer:
 
     def __init__(self, text="", tokenizer=None, device=None, prefix_token_ids=[]):

whisperlivekit/timed_objects.py

@@ -1,6 +1,6 @@
 from dataclasses import dataclass, field
-from typing import Optional, List, Union, Dict, Any
 from datetime import timedelta
+from typing import Any, Dict, List, Optional, Union
 
 PUNCTUATION_MARKS = {'.', '!', '?', '。', '！', '？'}
 

whisperlivekit/tokens_alignment.py

@@ -1,7 +1,9 @@
 from time import time
-from typing import Optional, List, Tuple, Union, Any
+from typing import Any, List, Optional, Tuple, Union
 
-from whisperlivekit.timed_objects import Line, SilentLine, ASRToken, SpeakerSegment, Silence, TimedText, Segment
+from whisperlivekit.timed_objects import (ASRToken, Line, Segment, Silence,
+                                          SilentLine, SpeakerSegment,
+                                          TimedText)
 
 
 class TokensAlignment:

whisperlivekit/warmup.py

@@ -7,6 +7,7 @@ def load_file(warmup_file=None, timeout=5):
     import os
     import tempfile
     import urllib.request
+
     import librosa
 
     if warmup_file == "":

whisperlivekit/web/web_interface.py

@@ -1,6 +1,6 @@
-import logging
-import importlib.resources as resources
 import base64
+import importlib.resources as resources
+import logging
 
 logger = logging.getLogger(__name__)
 
@@ -96,11 +96,13 @@ def get_inline_ui_html():
 
 if __name__ == '__main__':
     
+    import pathlib
+
+    import uvicorn
     from fastapi import FastAPI
     from fastapi.responses import HTMLResponse
-    import uvicorn
     from starlette.staticfiles import StaticFiles
-    import pathlib
+
     import whisperlivekit.web as webpkg
     
     app = FastAPI()    

whisperlivekit/whisper/__init__.py

@@ -4,15 +4,17 @@ import json
 import os
 import urllib
 import warnings
+from pathlib import Path
 from typing import Dict, List, Optional, Union
 
 import torch
-from tqdm import tqdm
-from pathlib import Path
 from torch import Tensor
+from tqdm import tqdm
 
-from whisperlivekit.whisper.audio import load_audio, log_mel_spectrogram, pad_or_trim
+from whisperlivekit.whisper.audio import (load_audio, log_mel_spectrogram,
-from whisperlivekit.whisper.decoding import DecodingOptions, DecodingResult, decode, detect_language
+                                          pad_or_trim)
+from whisperlivekit.whisper.decoding import (DecodingOptions, DecodingResult,
+                                             decode, detect_language)
 from whisperlivekit.whisper.model import ModelDimensions, Whisper
 from whisperlivekit.whisper.transcribe import transcribe
 from whisperlivekit.whisper.version import __version__

whisperlivekit/whisper/decoding.py

@@ -1,5 +1,6 @@
 from dataclasses import dataclass, field, replace
-from typing import TYPE_CHECKING, Dict, Iterable, List, Optional, Sequence, Tuple, Union
+from typing import (TYPE_CHECKING, Dict, Iterable, List, Optional, Sequence,
+                    Tuple, Union)
 
 import numpy as np
 import torch

whisperlivekit/whisper/transcribe.py

@@ -8,28 +8,13 @@ import numpy as np
 import torch
 import tqdm
 
-from .audio import (
+from .audio import (FRAMES_PER_SECOND, HOP_LENGTH, N_FRAMES, N_SAMPLES,
-    FRAMES_PER_SECOND,
+                    SAMPLE_RATE, log_mel_spectrogram, pad_or_trim)
-    HOP_LENGTH,
-    N_FRAMES,
-    N_SAMPLES,
-    SAMPLE_RATE,
-    log_mel_spectrogram,
-    pad_or_trim,
-)
 from .decoding import DecodingOptions, DecodingResult
 from .timing import add_word_timestamps
 from .tokenizer import LANGUAGES, TO_LANGUAGE_CODE, get_tokenizer
-from .utils import (
+from .utils import (exact_div, format_timestamp, get_end, get_writer,
-    exact_div,
+                    make_safe, optional_float, optional_int, str2bool)
-    format_timestamp,
-    get_end,
-    get_writer,
-    make_safe,
-    optional_float,
-    optional_int,
-    str2bool,
-)
 
 if TYPE_CHECKING:
     from .model import Whisper