Modernize: Gradio 6.29, Python 3.12, CPU torch, add ffmpeg, file paths

README.md

@@ -4,7 +4,8 @@ emoji: 📚
 colorFrom: indigo
 colorTo: green
 sdk: gradio
-sdk_version: 4.44.1
+sdk_version: 6.29.0
+python_version: "3.12"
 app_file: app.py
 pinned: false
 license: mit

app.py

@@ -6,11 +6,13 @@ model = whisper.load_model("base")
 
 def transcribe_audio(audio_file):
     # Check file size (e.g., 25MB limit)
-    if os.path.getsize(audio_file.name) > 25 * 1024 * 1024:
+    if not audio_file:
+        return "Error: please upload a file.", None
+    if os.path.getsize(audio_file) > 25 * 1024 * 1024:
         return "Error: File size exceeds 25MB limit.", None
 
-    result = model.transcribe(audio_file.name)
-    output_filename = os.path.splitext(os.path.basename(audio_file.name))[0] + ".txt"
+    result = model.transcribe(audio_file)
+    output_filename = os.path.splitext(os.path.basename(audio_file))[0] + ".txt"
     
     with open(output_filename, "w") as text_file:
         text_file.write(result["text"])
@@ -19,7 +21,6 @@ def transcribe_audio(audio_file):
 
 iface = gr.Interface(
     fn=transcribe_audio,
-    theme="Nymbo/Nymbo_Theme",
     inputs=gr.File(label="Upload Audio File (Max 25MB)"),
     outputs=[
         gr.Textbox(label="Transcription"),
@@ -29,4 +30,4 @@ iface = gr.Interface(
     description="Upload an audio file (WAV, MP3, etc.) up to 25MB to get its transcription. The transcript will be displayed and available for download. Please use responsibly."
 )
 
-iface.launch(share=True)
+iface.launch(theme="Nymbo/Nymbo_Theme")

packages.txt

@@ -0,0 +1 @@
+ffmpeg

requirements.txt

@@ -1,3 +1,3 @@
-gradio
+--extra-index-url https://download.pytorch.org/whl/cpu
 openai-whisper
-torch
+torch