Update README.md

Update README.md

Update README.md

Update README.md

Attach LTX-2.x Community License (contest terms 15)

Update scripts/dataset/README.md

scripts/dataset: add wetdry.py

scripts/dataset: add check_clips.py

scripts/dataset: add check_limiting.py

scripts/dataset: add verify_dataset.py

scripts/dataset: add make_manifest_v5.py

scripts/dataset: add build_eurorack_dry.py

scripts/dataset: add build_v5_sfx.py

scripts/dataset: add build_v5_outdoor.py

scripts/dataset: add build_v5_rooms.py

scripts/dataset: document the pair-building rules

scripts: add WORKING_CONFIG.md

scripts: add build_flf25.py

scripts: add trainer_setup.sh

scripts: add measure_timing.py

scripts: add build_heldout_refs.py

scripts: add build_sweep_manifest.py

scripts: add ltx_a2a_generate_multi.py

scripts: add README.md

examples: add techno-medium-room.mp3 (cyber_stand, ck11500)

examples: add techno-empty-club.mp3 (cyber_stand, ck11500)

examples: add techno-dual-delay.mp3 (cyber_stand, ck11500)

Update image caption

Use the scene-view screenshot

Embed the control-surface screenshot

Add control-surface screenshot

Restructure: short usage-focused card, full timing table and training config collapsed

Fold clip spec into the table; drop the duplicated prose

Checkpoint chosen by ear from test renders

Trim the card; correct the provenance wording

Tighten the video section to the proven chain

Correct the video-chain note: Comfy has the audio latent pieces; two-pass is the validated route, not the only possible one

Clarify: two-pass video chain is a ComfyUI tooling gap, not a model limitation

Lead the card with the spatial-awareness positioning; add dataset figures

AKUSPACE LTX-2.5 Audio LoRA v0.5 — checkpoint 11500, card, config, six A/B examples

initial commit