add output_dir arg

app.py

@@ -27,25 +27,25 @@ class App:
 
         if whisper_type in ["faster_whisper", "faster-whisper", "fasterwhisper"]:
             whisper_inf = FasterWhisperInference(
-                model_dir=self.args.faster_whisper_model_dir
+                model_dir=self.args.faster_whisper_model_dir,
+                output_dir=self.args.output_dir
             )
-            whisper_inf.model_dir = self.args.faster_whisper_model_dir
         elif whisper_type in ["whisper"]:
             whisper_inf = WhisperInference(
-                model_dir=self.args.whisper_model_dir
+                model_dir=self.args.whisper_model_dir,
+                output_dir=self.args.output_dir
             )
-            whisper_inf.model_dir = self.args.whisper_model_dir
         elif whisper_type in ["insanely_fast_whisper", "insanely-fast-whisper", "insanelyfastwhisper",
                               "insanely_faster_whisper", "insanely-faster-whisper", "insanelyfasterwhisper"]:
             whisper_inf = InsanelyFastWhisperInference(
-                model_dir=self.args.insanely_fast_whisper_model_dir
+                model_dir=self.args.insanely_fast_whisper_model_dir,
+                output_dir=self.args.output_dir
             )
-            whisper_inf.model_dir = self.args.insanely_fast_whisper_model_dir
         else:
             whisper_inf = FasterWhisperInference(
-                model_dir=self.args.faster_whisper_model_dir
+                model_dir=self.args.faster_whisper_model_dir,
+                output_dir=self.args.output_dir
             )
-            whisper_inf.model_dir = self.args.faster_whisper_model_dir
         return whisper_inf
 
     @staticmethod
@@ -387,6 +387,7 @@ parser.add_argument('--api_open', type=bool, default=False, nargs='?', const=Tru
 parser.add_argument('--whisper_model_dir', type=str, default=os.path.join("models", "Whisper"), help='Directory path of the whisper model')
 parser.add_argument('--faster_whisper_model_dir', type=str, default=os.path.join("models", "Whisper", "faster-whisper"), help='Directory path of the faster-whisper model')
 parser.add_argument('--insanely_fast_whisper_model_dir', type=str, default=os.path.join("models", "Whisper", "insanely-fast-whisper"), help='Directory path of the insanely-fast-whisper model')
+parser.add_argument('--output_dir', type=str, default=os.path.join("outputs"), help='Directory path of the outputs')
 _args = parser.parse_args()
 
 if __name__ == "__main__":

modules/faster_whisper_inference.py

@@ -18,10 +18,12 @@ os.environ['KMP_DUPLICATE_LIB_OK'] = 'True'
 
 class FasterWhisperInference(WhisperBase):
     def __init__(self,
-                 model_dir: str
+                 model_dir: str,
+                 output_dir: str
                  ):
         super().__init__(
-            model_dir=model_dir
+            model_dir=model_dir,
+            output_dir=output_dir
         )
         self.model_paths = self.get_model_paths()
         self.available_models = self.model_paths.keys()

modules/insanely_fast_whisper_inference.py

@@ -16,10 +16,12 @@ from modules.whisper_base import WhisperBase
 
 class InsanelyFastWhisperInference(WhisperBase):
     def __init__(self,
-                 model_dir: str
+                 model_dir: str,
+                 output_dir: str
                  ):
         super().__init__(
-            model_dir=model_dir
+            model_dir=model_dir,
+            output_dir=output_dir
         )
         openai_models = whisper.available_models()
         distil_models = ["distil-large-v2", "distil-large-v3", "distil-medium.en", "distil-small.en"]

modules/whisper_Inference.py

@@ -12,10 +12,12 @@ from modules.whisper_parameter import *
 
 class WhisperInference(WhisperBase):
     def __init__(self,
-                 model_dir: str
+                 model_dir: str,
+                 output_dir: str
                  ):
         super().__init__(
-            model_dir=model_dir
+            model_dir=model_dir,
+            output_dir=output_dir
         )
 
     def transcribe(self,

modules/whisper_base.py

@@ -15,10 +15,14 @@ from modules.whisper_parameter import *
 
 class WhisperBase(ABC):
     def __init__(self,
-                 model_dir: str):
+                 model_dir: str,
+                 output_dir: str
+                 ):
         self.model = None
         self.current_model_size = None
         self.model_dir = model_dir
+        self.output_dir = output_dir
+        os.makedirs(self.output_dir, exist_ok=True)
         os.makedirs(self.model_dir, exist_ok=True)
         self.available_models = whisper.available_models()
         self.available_langs = sorted(list(whisper.tokenizer.LANGUAGES.values()))
@@ -88,7 +92,8 @@ class WhisperBase(ABC):
                     file_name=file_name,
                     transcribed_segments=transcribed_segments,
                     add_timestamp=add_timestamp,
-                    file_format=file_format
+                    file_format=file_format,
+                    output_dir=self.output_dir
                 )
                 files_info[file_name] = {"subtitle": subtitle, "time_for_task": time_for_task, "path": file_path}
 
@@ -152,7 +157,8 @@ class WhisperBase(ABC):
                 file_name="Mic",
                 transcribed_segments=transcribed_segments,
                 add_timestamp=True,
-                file_format=file_format
+                file_format=file_format,
+                output_dir=self.output_dir
             )
 
             result_str = f"Done in {self.format_time(time_for_task)}! Subtitle file is in the outputs folder.\n\n{subtitle}"
@@ -211,7 +217,8 @@ class WhisperBase(ABC):
                 file_name=file_name,
                 transcribed_segments=transcribed_segments,
                 add_timestamp=add_timestamp,
-                file_format=file_format
+                file_format=file_format,
+                output_dir=self.output_dir
             )
             result_str = f"Done in {self.format_time(time_for_task)}! Subtitle file is in the outputs folder.\n\n{subtitle}"
 
@@ -237,6 +244,7 @@ class WhisperBase(ABC):
                                 transcribed_segments: list,
                                 add_timestamp: bool,
                                 file_format: str,
+                                output_dir: str
                                 ) -> str:
         """
         Writes subtitle file
@@ -251,6 +259,8 @@ class WhisperBase(ABC):
             Determines whether to add a timestamp to the end of the filename.
         file_format: str
             File format to write. Supported formats: [SRT, WebVTT, txt]
+        output_dir: str
+            Directory path of the output
 
         Returns
         ----------
@@ -261,9 +271,9 @@ class WhisperBase(ABC):
         """
         timestamp = datetime.now().strftime("%m%d%H%M%S")
         if add_timestamp:
-            output_path = os.path.join("outputs", f"{file_name}-{timestamp}")
+            output_path = os.path.join(output_dir, f"{file_name}-{timestamp}")
         else:
-            output_path = os.path.join("outputs", f"{file_name}")
+            output_path = os.path.join(output_dir, f"{file_name}")
 
         if file_format == "SRT":
             content = get_srt(transcribed_segments)