Add explicit Target Duration slider (0–60s) + gen_duration kwarg

Move LTX acknowledgement to top of Space README, position DramaBox as Resemble AI's product

Revert: keep DramaBox naming (rebrand reverted per CEO)

Rebrand: DramaBox → LTX-2.3-Voice

Patch end-of-clip silence prior at latent frame 513 for >20s outputs

Drop video_connector + video_aggregate_embed BEFORE .to(device)

Use richer duration estimator in warm server (sentence + non-verbal aware)

Warm-load TTSServer at module level (IndexTTS pattern)

Add 8 voice references + click-to-generate Examples table

Replace example dropdown with click-to-generate gr.Examples table

Default Gradio server to port 7860 (HF Spaces gateway expects this)

Pin pydantic==2.10.6 — root cause of the recurring bool-iter crash

Disable SSR + analytics + auto-API for ZeroGPU compat

Bump to gradio 5.7.1; drop pydantic/starlette/fastapi 4.x compat pins

Pin starlette<1.0 + fastapi<0.116 for gradio 4.44.1 TemplateResponse compat

Pin pydantic==2.10.6 to fix gradio_client schema bool TypeError

Pin SDK to 4.44.1 + monkey-patch gradio_client schema bool TypeError

Bump gradio to 5.x to fix gradio_client schema TypeError on bool

Pin torch/torchaudio==2.4.0 to fix ZeroGPU libcudart.so.13 mismatch

Refactor for ZeroGPU: lazy TTSServer load + @spaces.GPU decorator

Fetch unsloth bnb-4bit Gemma + Dramabox checkpoints from HF on startup

Pin huggingface_hub<1.0 (gradio 4.44.1 still imports HfFolder)

Restore Space README with sdk frontmatter

DramaBox Space — initial app + vendored ltx2

DramaBox Space README

initial commit