luke9705 commited on
Commit
c4a371d
·
1 Parent(s): 2ddbac2

Enhance Gradio interface: Update title to 'Scriptura: A MultiAgent System for Screenplay Creation and Editing' and add detailed description for improved user understanding.

Browse files
Files changed (1) hide show
  1. app.py +21 -2
app.py CHANGED
@@ -13,6 +13,7 @@ import openai
13
  from openai import OpenAI
14
  import pdfplumber
15
  import numpy as np
 
16
 
17
 
18
  ## utilties and class definition
@@ -300,19 +301,37 @@ def initialize_agent():
300
  return agent
301
 
302
  ## gradio interface
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
303
  global agent
304
  agent = initialize_agent()
305
  demo = gr.ChatInterface(
306
  fn=respond,
307
  type='messages',
308
  multimodal=True,
309
- title='MultiAgent System for Screenplay Creation and Editing',
 
310
  show_progress='full',
311
  fill_height=True,
312
  fill_width=True,
313
  save_history=True,
314
  autoscroll=True,
315
- #css = css_snippet,
316
  additional_inputs=[
317
  gr.Checkbox(value=False, label="Web Search",
318
  info="Enable web search to find information online. If disabled, the agent will only use the provided files and images.",
 
13
  from openai import OpenAI
14
  import pdfplumber
15
  import numpy as np
16
+ import textwrap
17
 
18
 
19
  ## utilties and class definition
 
301
  return agent
302
 
303
  ## gradio interface
304
+ description = textwrap.dedent("""**Scriptura** is a multi-agent AI framework based on HF-SmolAgents that streamlines the creation of screenplays, storyboards,
305
+ and soundtracks by automating the stages of analysis, summarization, and multimodal enrichment, freeing authors to focus on pure creativity.
306
+ At its heart:
307
+ - **Qwen3-32B** serves as the primary orchestrating agent, coordinating workflows and managing high-level reasoning across the system.
308
+ - **Gemma-3-27B-IT** acts as a specialized assistant for multimodal tasks, supporting both text and audio inputs to refine narrative elements and prepare them for downstream generation.
309
+
310
+ For media generation, Scriptura integrates:
311
+ - **MusicGen** models (per the AudioCraft MusicGen specification), deployed via Hugging Face Spaces,
312
+ enabling the agent to produce original soundtracks and sound effects from text prompts or combined text + audio samples.
313
+ - **FLUX (black-forest-labs/FLUX.1-dev)** for on-the-fly image creation—ideal for storyboards, concept art, and
314
+ visual references that seamlessly tie into the narrative flow.
315
+
316
+ Optionally, Scriptura can query external sources (e.g., via a DuckDuckGo API integration) to pull in reference scripts, sound samples, or research materials,
317
+ ensuring that every draft is not only creatively rich but also contextually informed.
318
+
319
+ For more information: [README.md](https://huggingface.co/spaces/Lab9705/MultiAgent_System_for_Screenplay_Creation/blob/main/README.md)
320
+ """)
321
+
322
  global agent
323
  agent = initialize_agent()
324
  demo = gr.ChatInterface(
325
  fn=respond,
326
  type='messages',
327
  multimodal=True,
328
+ title='Scriptura: A MultiAgent System for Screenplay Creation and Editing',
329
+ description=description,
330
  show_progress='full',
331
  fill_height=True,
332
  fill_width=True,
333
  save_history=True,
334
  autoscroll=True,
 
335
  additional_inputs=[
336
  gr.Checkbox(value=False, label="Web Search",
337
  info="Enable web search to find information online. If disabled, the agent will only use the provided files and images.",