feat: add multiple audio files handling

Author: massarin

.gitignore

@@ -1,3 +1,4 @@
+# Artifacts
 __pycache__
 .ipynb_checkpoints/
 .DS_Store
@@ -14,13 +15,14 @@ venv
 build/
 dist/
 _version.py
+
+# Media and output
 *.txt
-!testing_audio.txt
 *.srt
-!testing_audio.srt
 *.wav
-!testing_audio.wav
 *.mp3
-!testing_audio.mp3
 *.m4a
-!testing_audio.m4a
+!test0*
+
+# Deprecated
+requirements.txt

README.md

@@ -37,8 +37,8 @@ execute()
 
 ### Python API
 ```python
-from ghe_transcribe.core import transcribe_core
-result = transcribe_core("path/to/audio.mp3")
+from ghe_transcribe.core import transcribe
+result = transcribe("path/to/audio.mp3")
 ```
 
 ### Command Line

media/testing_audio.mp3 media/test01.mp3media/testing_audio.mp3 renamed to media/test01.mp3

@@ -0,0 +1,27 @@
+1
+00:00:00,640 --> 00:00:08,960
+S00: So hi Fiona, we're here to record an audio that I will use for testing.
+
+2
+00:00:10,400 --> 00:00:17,600
+S01: Hi Nicolo, thanks so much for the honor to invite me to help you on this project.
+
+3
+00:00:18,240 --> 00:00:34,160
+S00: You're welcome, you were the first person that came to mind. So we're here to ask you a couple of questions until we reach minute one. So what do you do in life Fiona?
+
+4
+00:00:34,160 --> 00:00:44,400
+S02: Good question, good question. So at the moment I do an internship, it's going great, everything I do is productive, will be published probably by tomorrow.
+
+5
+00:00:46,400 --> 00:00:53,280
+S02: Yeah, just doing the grind, just yeah. Of course. Do you have plans for lunch?
+
+6
+00:00:53,280 --> 00:00:56,560
+S02: Plans for lunch? Yes, actually I do, I'm meeting a friend.
+
+7
+00:00:56,560 --> 00:01:02,400
+S02: If you want, you're welcome to join. Perfect, thank you very much.

media/test02.m4a

@@ -0,0 +1,7 @@
+S00: [00:00:00]  So hi Fiona, we're here to record an audio that I will use for testing.
+S01: [00:00:10]  Hi Nicolo, thanks so much for the honor to invite me to help you on this project.
+S00: [00:00:18]  You're welcome, you were the first person that came to mind. So we're here to ask you a couple of questions until we reach minute one. So what do you do in life Fiona?
+S02: [00:00:34]  Good question, good question. So at the moment I do an internship, it's going great, everything I do is productive, will be published probably by tomorrow.
+S02: [00:00:46]  Yeah, just doing the grind, just yeah. Of course. Do you have plans for lunch?
+S02: [00:00:53]  Plans for lunch? Yes, actually I do, I'm meeting a friend.
+S02: [00:00:56]  If you want, you're welcome to join. Perfect, thank you very much.

media/testing_audio.wav

@@ -17,8 +17,8 @@
     ComputeTypeChoice,  # Enum for compute type choices
     DeviceChoice,  # Enum for device choices
     WhisperModelChoice,  # Enum for Whisper model choices
+    transcribe,
     transcribe_config,  # Default configuration
-    transcribe_core,
 )
 from ghe_transcribe.utils import log_hf_authentication_error, save_uploaded_file
 
@@ -146,7 +146,7 @@ def _setup_ui(self):
         """Defines all the UI widgets and their initial layout."""
         # Basic Options
         self.audio_uploader = widgets.FileUpload(
-            multiple=False,
+            multiple=True,
             description="Upload Audio",
             # accept=".wav, .m4a, .mp3, .flac, .ogg",  # Specify accepted audio formats
             layout=self.common_widget_layout,
@@ -369,21 +369,26 @@ def _on_run_button_click(self, b):
                 return
 
             try:
-                # Handle uploaded file
-                file_metadata = self.audio_uploader.value[0]
-                uploaded_file_name = file_metadata["name"]
-                uploaded_content_bytes = file_metadata["content"].tobytes()
-
-                # Save uploaded file using modern path handling
-                audio_file_path = save_uploaded_file(
-                    uploaded_file_name, uploaded_content_bytes
-                )
-                logger.info(f"Uploaded audio saved to: {audio_file_path}")
-                print(f"Uploaded audio saved to: {audio_file_path}")
+                # Handle uploaded files (single or multiple)
+                uploaded_files = []
+                for file_metadata in self.audio_uploader.value:
+                    uploaded_file_name = file_metadata["name"]
+                    uploaded_content_bytes = file_metadata["content"].tobytes()
+
+                    # Save uploaded file using modern path handling
+                    audio_file_path = save_uploaded_file(
+                        uploaded_file_name, uploaded_content_bytes
+                    )
+                    uploaded_files.append(str(audio_file_path))
+                    logger.info(f"Uploaded audio saved to: {audio_file_path}")
+                    print(f"Uploaded audio saved to: {audio_file_path}")
+
+                # Determine if single file or multiple files
+                files_input = uploaded_files[0] if len(uploaded_files) == 1 else uploaded_files
 
                 # Prepare arguments for transcribe
                 kwargs = {
-                    "file": str(audio_file_path),
+                    "files": files_input,
                     "trim": self.trim_input.value
                     if self.trim_input.value > 0
                     else None,
@@ -413,7 +418,7 @@ def _on_run_button_click(self, b):
                     kwargs["max_speakers"] = self.max_speakers_input.value
 
                 # Call the ghe_transcribe function
-                transcribe_core(**kwargs)
+                transcribe(**kwargs)
 
             except Exception as e:
                 logger.error(f"An unexpected error occurred: {e}", exc_info=True)