Update app.py

app.py

@@ -16,6 +16,7 @@ B_KEY = os.getenv("B_KEY")
 
 # URLs
 API_URL = os.getenv("API_URL")
+UPLOAD_URL = os.getenv("UPLOAD_URL")
 
 def get_voices():
     # OpenAI TTS voices
@@ -48,18 +49,23 @@ def text_to_speech(voice, text):
     
     return response.content
 
-def lipsync_api_call(video_url, audio_content):
+def upload_file(file_content, file_name):
+    files = {'fileToUpload': (file_name, file_content)}
+    data = {'reqtype': 'fileupload'}
+    response = requests.post(UPLOAD_URL, files=files, data=data)
+    
+    if response.status_code == 200:
+        return response.text.strip()
+    return None
+
+def lipsync_api_call(video_url, audio_url):
     headers = {
         "Content-Type": "application/json",
         "x-api-key": B_KEY
     }
     
-    # Create a multipart form-data request
-    files = {
-        'audio': ('audio.mp3', audio_content, 'audio/mpeg')
-    }
-    
     data = {
+        "audioUrl": audio_url,
         "videoUrl": video_url,
         "maxCredits": 1000,
         "model": "sync-1.7.1-beta",
@@ -68,7 +74,7 @@ def lipsync_api_call(video_url, audio_content):
         "synergizerStrength": 1
     }
     
-    response = requests.post(API_URL, headers=headers, data=data, files=files)
+    response = requests.post(API_URL, headers=headers, json=data)
     return response.json()
 
 def check_job_status(job_id):
@@ -88,7 +94,6 @@ def check_job_status(job_id):
     return None
 
 def get_media_duration(file_path):
-    # Fetch media duration using ffprobe
     cmd = ['ffprobe', '-v', 'error', '-show_entries', 'format=duration', '-of', 'default=noprint_wrappers=1:nokey=1', file_path]
     result = subprocess.run(cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
     return float(result.stdout.strip())
@@ -130,16 +135,20 @@ def combine_audio_video(video_path, audio_content, output_path):
 
 def process_video(voice, video_url, text, progress=gr.Progress()):
     session_id = str(uuid.uuid4())  # Generate a unique session ID
+    
     progress(0, desc="Generating speech...")
     audio_content = text_to_speech(voice, text)
     if not audio_content:
         return None, "Failed to generate speech audio."
     
-    progress(0.2, desc="Processing video...")
+    progress(0.2, desc="Uploading audio...")
+    audio_url = upload_file(audio_content, f"audio_{session_id}.mp3")
+    if not audio_url:
+        return None, "Failed to upload audio file."
     
     try:
         progress(0.4, desc="Initiating lipsync...")
-        job_data = lipsync_api_call(video_url, audio_content)
+        job_data = lipsync_api_call(video_url, audio_url)
         
         if "error" in job_data or "message" in job_data:
             raise Exception(job_data.get("error", job_data.get("message", "Unknown error")))