Update spaces.py

spaces.py

@@ -159,7 +159,7 @@ def create_video_clip(image, duration, target_resolution=(1920, 1080)):
     return clip.set_duration(duration)
 
 def process_message(args):
-    i, message, logo_image, voice_ids = args
+    i, message, logo_image, voice_ids, male_stability, male_style = args
     voice_id = voice_ids[i % len(voice_ids)]
 
     if i % len(voice_ids) == 0:
@@ -168,8 +168,8 @@ def process_message(args):
         style = 0
     else:
         text_color = "#FFFFFF"
-        stability = 0.65
-        style = 0.35
+        stability = male_stability
+        style = male_style
 
     try:
         audio_content = generate_speech(message, voice_id, stability=stability, style=style)
@@ -197,7 +197,7 @@ def process_message(args):
         print(f"Error processing message {i+1}: {e}")
         return (None, None, None)
 
-def generate_conversation_video(messages, voice_ids, logo_url):
+def generate_conversation_video(messages, voice_ids, logo_url, male_stability, male_style):
     logo_image = download_and_convert_svg_to_png(logo_url)
     if logo_image is None:
         return None
@@ -206,7 +206,7 @@ def generate_conversation_video(messages, voice_ids, logo_url):
     audio_clips = []
     temp_audio_paths = []
 
-    args = [(i, message, logo_image, voice_ids) for i, message in enumerate(messages)]
+    args = [(i, message, logo_image, voice_ids, male_stability, male_style) for i, message in enumerate(messages)]
     max_workers = 5
 
     with concurrent.futures.ThreadPoolExecutor(max_workers=max_workers) as executor:
@@ -259,7 +259,7 @@ def generate_conversation_video(messages, voice_ids, logo_url):
 
     return temp_video_path
 
-def generate_video(description):
+def generate_video(description, male_stability=0.65, male_style=0.35):
     voice_ids = [
         "cgSgspJ2msm6clMCkdW9",  # First speaker
         "3Niy6MUaDzcs7Liw7dFs"   # Second speaker
@@ -267,19 +267,35 @@ def generate_video(description):
     logo_url = "https://opencall.ai/images/logo-symbol.svg"
     
     messages = get_convo_list(description)
-    video_path = generate_conversation_video(messages, voice_ids, logo_url)
+    video_path = generate_conversation_video(messages, voice_ids, logo_url, male_stability, male_style)
     
     return video_path
 
 # Create Gradio interface
 iface = gr.Interface(
     fn=generate_video,
-    inputs=gr.Textbox(
-        label="Enter conversation description",
-        lines=5,
-        placeholder="Describe the conversation you want to generate...",
-        info="You can be specific about the number of turns, tone, and content of the conversation"
-    ),
+    inputs=[
+        gr.Textbox(
+            label="Enter conversation description",
+            lines=5,
+            placeholder="Describe the conversation you want to generate...",
+            info="You can be specific about the number of turns, tone, and content of the conversation"
+        ),
+        gr.Slider(
+            minimum=0.1,
+            maximum=1.0,
+            value=0.65,
+            label="Male Voice Stability",
+            info="Controls the consistency of the male voice (default: 0.65)"
+        ),
+        gr.Slider(
+            minimum=0.1,
+            maximum=1.0,
+            value=0.35,
+            label="Male Voice Style",
+            info="Controls the expressiveness of the male voice (default: 0.35)"
+        )
+    ],
     outputs=gr.Video(label="Generated Video"),
     title="AI Conversation Video Generator",
     description="Generate a video conversation between two speakers based on your description."