Commit History

  1. Add professional README with animated SVG banner, features table, and architecture diagram

    ShebMichel
    8dcd60e
  2. Add retry logic for vision model (handles 500/504 timeouts)

    ShebMichel
    e61ec84
  3. Fix: replace non-existent en-US-DavisNeural with ChristopherNeural

    ShebMichel
    51fa180
  4. Try Qwen2.5-VL-72B-Instruct for vision

    ShebMichel
    d823dea
  5. Switch vision model to Qwen/Qwen2.5-VL-7B-Instruct (supported by inference providers)

    ShebMichel
    40773b2
  6. Fix vision model: switch to Llama-3.2-11B-Vision + resize images to 1024px max

    ShebMichel
    a5f0ded
  7. Add error handling with clear error messages for each pipeline stage

    ShebMichel
    6f678cc
  8. Add Chair/moderator who introduces the topic and speakers

    ShebMichel
    77a5ef5
  9. Complete redesign: Small Talk style side-by-side robot panels with captions, embedded audio auto-play via iframe srcdoc

    ShebMichel
    a1491bc
  10. Fix robot sync: use polling watcher + hidden textbox for timeline data (scripts dont re-execute in gr.HTML updates)

    ShebMichel
    d16a5b3
  11. Fix autoplay: use inline HTML (no iframe) + sync to Gradio autoplay audio element

    ShebMichel
    2ba8d81

  1. Fix autoplay: add click-to-play overlay fallback and allow-autoplay on iframe

    ShebMichel
    a6d8ed5
  2. Auto-play audio in robot viewer with synced transcript log

    ShebMichel
    794a3ef
  3. Add animated robot avatars synced to speaker turns

    ShebMichel
    8eec23c

  1. Update edge-tts to >=7.2.5 to fix 403 token issue

    ShebMichel
    8b035c8
  2. Use CohereLabs/aya-vision-32b - confirmed available on HF inference

    ShebMichel
    1f38743
  3. Switch to Qwen models - available on HF inference without gating

    ShebMichel
    b657956
  4. Fix asyncio: use new_event_loop instead of nest_asyncio

    ShebMichel
    8b98e34
  5. Remove explicit provider - let auto-router select with valid HF token

    ShebMichel
    808cfd0
  6. Fix InferenceClient: explicitly use hf-inference provider

    ShebMichel
    8e36e22
  7. Fix huggingface-hub version to be compatible with gradio 6.19

    ShebMichel
    ef4c511
  8. Remove gradio pin - HF Spaces provides it via sdk_version

    ShebMichel
    fff0460
  9. Initial commit: GeoTalk AI Geoscience Debate Podcast

    ShebMichel
    5376294
  10. initial commit

    Verified signature
    0d6f6d3