refactored directory structure

app.py

@@ -1,15 +1,14 @@
-import gradio as gr
 import os
 import argparse
 
-from modules.whisper_Inference import WhisperInference
-from modules.faster_whisper_inference import FasterWhisperInference
-from modules.insanely_fast_whisper_inference import InsanelyFastWhisperInference
-from modules.nllb_inference import NLLBInference
+from modules.whisper.whisper_Inference import WhisperInference
+from modules.whisper.faster_whisper_inference import FasterWhisperInference
+from modules.whisper.insanely_fast_whisper_inference import InsanelyFastWhisperInference
+from modules.translation.nllb_inference import NLLBInference
 from ui.htmls import *
-from modules.youtube_manager import get_ytmetas
-from modules.deepl_api import DeepLAPI
-from modules.whisper_parameter import *
+from modules.utils.youtube_manager import get_ytmetas
+from modules.translation.deepl_api import DeepLAPI
+from modules.whisper.whisper_parameter import *
 
 
 class App:

modules/{diarizer.py → diarize/diarizer.py}

@@ -1,11 +1,10 @@
 import os
-import whisperx
 import torch
 from typing import List
 import time
 
-from modules.diarize_pipeline import DiarizationPipeline
-
+from modules.diarize.diarize_pipeline import DiarizationPipeline, assign_word_speakers
+from modules.diarize.audio_loader import load_audio
 
 class Diarizer:
     def __init__(self,
@@ -55,9 +54,9 @@ class Diarizer:
                 use_auth_token=use_auth_token
             )
 
-        audio = whisperx.load_audio(audio)
+        audio = load_audio(audio)
         diarization_segments = self.pipe(audio)
-        diarized_result = whisperx.assign_word_speakers(
+        diarized_result = assign_word_speakers(
             diarization_segments,
             {"segments": transcribed_result}
         )

modules/{deepl_api.py → translation/deepl_api.py}

@@ -4,7 +4,7 @@ import os
 from datetime import datetime
 import gradio as gr
 
-from modules.subtitle_manager import *
+from modules.utils.subtitle_manager import *
 
 """
 This is written with reference to the DeepL API documentation.
@@ -144,7 +144,7 @@ class DeepLAPI:
                 timestamp = datetime.now().strftime("%m%d%H%M%S")
 
                 file_name = file_name[:-9]
-                output_path = os.path.join(self.output_dir, "translations", f"{file_name}-{timestamp}.srt")
+                output_path = os.path.join(self.output_dir, "", f"{file_name}-{timestamp}.srt")
                 write_file(subtitle, output_path)
 
             elif file_ext == ".vtt":
@@ -164,7 +164,7 @@ class DeepLAPI:
                 timestamp = datetime.now().strftime("%m%d%H%M%S")
 
                 file_name = file_name[:-9]
-                output_path = os.path.join(self.output_dir, "translations", f"{file_name}-{timestamp}.vtt")
+                output_path = os.path.join(self.output_dir, "", f"{file_name}-{timestamp}.vtt")
 
                 write_file(subtitle, output_path)
 

modules/{nllb_inference.py → translation/nllb_inference.py}

@@ -2,7 +2,7 @@ from transformers import AutoTokenizer, AutoModelForSeq2SeqLM, pipeline
 import gradio as gr
 import os
 
-from modules.translation_base import TranslationBase
+from modules.translation.translation_base import TranslationBase
 
 
 class NLLBInference(TranslationBase):

modules/{translation_base.py → translation/translation_base.py}

@@ -5,8 +5,8 @@ from abc import ABC, abstractmethod
 from typing import List
 from datetime import datetime
 
-from modules.whisper_parameter import *
-from modules.subtitle_manager import *
+from modules.whisper.whisper_parameter import *
+from modules.utils.subtitle_manager import *
 
 
 class TranslationBase(ABC):
@@ -90,9 +90,9 @@ class TranslationBase(ABC):
 
                     timestamp = datetime.now().strftime("%m%d%H%M%S")
                     if add_timestamp:
-                        output_path = os.path.join("outputs", "translations", f"{file_name}-{timestamp}.srt")
+                        output_path = os.path.join("outputs", "", f"{file_name}-{timestamp}.srt")
                     else:
-                        output_path = os.path.join("outputs", "translations", f"{file_name}.srt")
+                        output_path = os.path.join("outputs", "", f"{file_name}.srt")
 
                 elif file_ext == ".vtt":
                     parsed_dicts = parse_vtt(file_path=file_path)
@@ -105,9 +105,9 @@ class TranslationBase(ABC):
 
                     timestamp = datetime.now().strftime("%m%d%H%M%S")
                     if add_timestamp:
-                        output_path = os.path.join(self.output_dir, "translations", f"{file_name}-{timestamp}.vtt")
+                        output_path = os.path.join(self.output_dir, "", f"{file_name}-{timestamp}.vtt")
                     else:
-                        output_path = os.path.join(self.output_dir, "translations", f"{file_name}.vtt")
+                        output_path = os.path.join(self.output_dir, "", f"{file_name}.vtt")
 
                 write_file(subtitle, output_path)
                 files_info[file_name] = subtitle

modules/{faster_whisper_inference.py → whisper/faster_whisper_inference.py}

@@ -9,8 +9,8 @@ import whisper
 import gradio as gr
 from argparse import Namespace
 
-from modules.whisper_parameter import *
-from modules.whisper_base import WhisperBase
+from modules.whisper.whisper_parameter import *
+from modules.whisper.whisper_base import WhisperBase
 
 
 class FasterWhisperInference(WhisperBase):

modules/{insanely_fast_whisper_inference.py → whisper/insanely_fast_whisper_inference.py}

@@ -11,8 +11,8 @@ import whisper
 from rich.progress import Progress, TimeElapsedColumn, BarColumn, TextColumn
 from argparse import Namespace
 
-from modules.whisper_parameter import *
-from modules.whisper_base import WhisperBase
+from modules.whisper.whisper_parameter import *
+from modules.whisper.whisper_base import WhisperBase
 
 
 class InsanelyFastWhisperInference(WhisperBase):

modules/{whisper_Inference.py → whisper/whisper_Inference.py}

@@ -1,14 +1,13 @@
 import whisper
 import gradio as gr
 import time
-import os
 from typing import BinaryIO, Union, Tuple, List
 import numpy as np
 import torch
 from argparse import Namespace
 
-from modules.whisper_base import WhisperBase
-from modules.whisper_parameter import *
+from modules.whisper.whisper_base import WhisperBase
+from modules.whisper.whisper_parameter import *
 
 
 class WhisperInference(WhisperBase):

modules/{whisper_base.py → whisper/whisper_base.py}

@@ -1,6 +1,5 @@
 import os
 import torch
-from typing import List
 import whisper
 import gradio as gr
 from abc import ABC, abstractmethod
@@ -8,12 +7,11 @@ from typing import BinaryIO, Union, Tuple, List
 import numpy as np
 from datetime import datetime
 from argparse import Namespace
-import time
 
-from modules.subtitle_manager import get_srt, get_vtt, get_txt, write_file, safe_filename
-from modules.youtube_manager import get_ytdata, get_ytaudio
-from modules.whisper_parameter import *
-from modules.diarizer import Diarizer
+from modules.utils.subtitle_manager import get_srt, get_vtt, get_txt, write_file, safe_filename
+from modules.utils.youtube_manager import get_ytdata, get_ytaudio
+from modules.whisper.whisper_parameter import *
+from modules.diarize.diarizer import Diarizer
 
 
 class WhisperBase(ABC):