Added video input capability

app.py

@@ -1,5 +1,6 @@
 import os
 import urllib.request
+import subprocess
 from dataclasses import dataclass
 from time import sleep
 from typing import Dict, List, Generator
@@ -27,6 +28,24 @@ class MockInterviewer:
         self._assistant_id_cache: Dict[Config, str] = {}
         self.clear_thread()
 
+    def convert_webm_to_mp3(input_webm, output_mp3):
+        command = [
+            'ffmpeg',
+            '-i', input_webm,  # Input file
+            '-vn',             # No video (remove video stream)
+            '-ab', '160k',     # Audio bitrate
+            '-ar', '44100',    # Audio sample rate
+            '-y',              # Overwrite output file if it exists
+            '-f', 'mp3',       # Output format
+            output_mp3        # Output file
+        ]
+        try:
+            subprocess.run(command, check=True)
+            print(f"File converted successfully and saved as {output_mp3}")
+        except subprocess.CalledProcessError as e:
+            print(f"An error occurred while converting the file: {e}")
+
+
     def chat_with_text(
         self, 
         message: Dict, 
@@ -44,7 +63,7 @@ class MockInterviewer:
 
     def chat_with_audio(
         self, 
-        audio: str,
+        video: str,
         job_role: str, 
         company: str, 
         job_description: str,
@@ -53,6 +72,8 @@ class MockInterviewer:
         situational_count: int, 
         case_count: int
     ) -> str:
+        audio = 'temp_audio.mp3'
+        MockInterviewer.convert_webm_to_mp3(video,audio)
         with open(audio, 'rb') as audio_file:
             transcriptions = self._client.audio.transcriptions.create(
                 model='whisper-1',
@@ -62,7 +83,7 @@ class MockInterviewer:
         config = Config(job_role, company, job_description, behavioral_count, technical_count, situational_count, case_count)
         response = self._chat(transcriptions.text, config)
         return [(transcriptions.text, response)]
-    
+
     def clear_thread(self) -> None:
         print('Initializing new thread')
         self._thread = self._client.beta.threads.create() 
@@ -191,9 +212,9 @@ with gr.Blocks(theme=theme) as demo:
             chat_interface.load(mock_interviewer.clear_thread)
             chat_interface.clear_btn.click(mock_interviewer.clear_thread)
 
-            audio = gr.Audio(sources=['microphone'], type='filepath', editable=False)
-            audio.stop_recording(fn=mock_interviewer.chat_with_audio,
-                                inputs=[audio, job_role, company, job_description, behavioral_count, technical_count, situational_count, case_count],
+            video = gr.Video(sources='webcam', include_audio=True)
+            video.stop_recording(fn=mock_interviewer.chat_with_audio,
+                                inputs=[video, job_role, company, job_description, behavioral_count, technical_count, situational_count, case_count],
                                 outputs=[chat_interface.chatbot],
                                 api_name=False)