Merge pull request #195 from jhj0517/feature/local-input

app.py

@@ -81,8 +81,13 @@ class App:
                     gr.Markdown(MARKDOWN, elem_id="md_project")
             with gr.Tabs():
                 with gr.TabItem("File"):  # tab1
-                    with gr.Row():
+                    with gr.Column():
                         input_file = gr.Files(type="filepath", label="Upload File here")
+                        tb_input_folder = gr.Textbox(label="Input Folder Path (Optional)",
+                                                     info="Optional: Specify the folder path where the input files are located, if you prefer to use local files instead of uploading them" 
+                                                          " Leave this field empty if you do not wish to use a local path.",
+                                                     visible=self.args.colab,
+                                                     value="")
                     with gr.Row():
                         dd_model = gr.Dropdown(choices=self.whisper_inf.available_models, value="large-v2",
                                                label="Model")
@@ -128,7 +133,7 @@ class App:
                         files_subtitles = gr.Files(label="Downloadable output file", scale=3, interactive=False)
                         btn_openfolder = gr.Button('📂', scale=1)
 
-                    params = [input_file, dd_file_format, cb_timestamp]
+                    params = [input_file, tb_input_folder, dd_file_format, cb_timestamp]
                     whisper_params = WhisperParameters(model_size=dd_model,
                                                        lang=dd_lang,
                                                        is_translate=cb_translate,

modules/utils/files_manager.py

@@ -0,0 +1,39 @@
+import os
+import fnmatch
+
+from gradio.utils import NamedString
+
+
+def get_media_files(folder_path, include_sub_directory=False):
+    video_extensions = ['*.mp4', '*.mkv', '*.flv', '*.avi', '*.mov', '*.wmv']
+    audio_extensions = ['*.mp3', '*.wav', '*.aac', '*.flac', '*.ogg', '*.m4a']
+    media_extensions = video_extensions + audio_extensions
+
+    media_files = []
+
+    if include_sub_directory:
+        for root, _, files in os.walk(folder_path):
+            for extension in media_extensions:
+                media_files.extend(
+                    os.path.join(root, file) for file in fnmatch.filter(files, extension)
+                    if os.path.exists(os.path.join(root, file))
+                )
+    else:
+        for extension in media_extensions:
+            media_files.extend(
+                os.path.join(folder_path, file) for file in fnmatch.filter(os.listdir(folder_path), extension)
+                if os.path.isfile(os.path.join(folder_path, file)) and os.path.exists(os.path.join(folder_path, file))
+            )
+
+    return media_files
+
+
+def format_gradio_files(files: list):
+    if not files:
+        return files
+
+    gradio_files = []
+    for file in files:
+        gradio_files.append(NamedString(file))
+    return gradio_files
+

modules/whisper/whisper_base.py

@@ -12,6 +12,7 @@ from dataclasses import astuple
 
 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.utils.files_manager import get_media_files, format_gradio_files
 from modules.whisper.whisper_parameter import *
 from modules.diarize.diarizer import Diarizer
 from modules.vad.silero_vad import SileroVAD
@@ -123,6 +124,7 @@ class WhisperBase(ABC):
 
     def transcribe_file(self,
                         files: list,
+                        input_folder_path: str,
                         file_format: str,
                         add_timestamp: bool,
                         progress=gr.Progress(),
@@ -135,6 +137,9 @@ class WhisperBase(ABC):
         ----------
         files: list
             List of files to transcribe from gr.Files()
+        input_folder_path: str
+            Input folder path to transcribe from gr.Textbox(). If this is provided, `files` will be ignored and
+            this will be used instead.
         file_format: str
             Subtitle File format to write from gr.Dropdown(). Supported format: [SRT, WebVTT, txt]
         add_timestamp: bool
@@ -152,6 +157,10 @@ class WhisperBase(ABC):
             Output file path to return to gr.Files()
         """
         try:
+            if input_folder_path:
+                files = get_media_files(input_folder_path)
+                files = format_gradio_files(files)
+
             files_info = {}
             for file in files:
                 transcribed_segments, time_for_task = self.run(