nodes/havvk-ComfyUI_AIIA
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
671 Commits 1 Branch 0 Tags
c936887aedf6cf9be166325281db7336e494b120
Commit Graph
176 Commits
Author SHA1 Message Date
Hawk Lee c936887aed fix(ditto): Restore EMA Logic! Wire mouth_smoothing to MotionStitch instead of Audio2Motion v1.9.46 2026-01-22 22:28:08 +08:00
Hawk Lee d937f688e3 feat(ditto): Add 'Custom (Manual)' option to resolve conflict with smo_k_d input v1.9.45 2026-01-22 22:18:05 +08:00
Hawk Lee 4125bbce45 feat(ditto): Upgrade mouth_smoothing to Float Blend (0.0/0.3/0.5/0.7) v1.9.44 2026-01-22 22:12:56 +08:00
Hawk Lee aeb2ad027a fix(ditto): Wire up mouth_smoothing parameter (was ignored) v1.9.43 2026-01-22 22:08:32 +08:00
Hawk Lee d4d8103f9f feat(ditto): Implement Auto-Disk Mode (Graph-Aware) v1.9.42; Clean up pose_stabilization 2026-01-22 21:58:56 +08:00
Hawk Lee eb2dfb065c fix(ditto): Rollback to v1.9.31 + Add Vitality Noise Injection (v1.9.41) 2026-01-22 21:32:11 +08:00
Hawk Lee db6efd07c9 fix(video): Resolve UnboundLocalError and document zero-perf-loss Disk Mode 2026-01-22 19:35:28 +08:00
Hawk Lee e7054c2476 fix: Robust filename pattern detection & logging for VideoCombine v1.9.30 2026-01-22 19:18:36 +08:00
Hawk Lee b0a2bf38fa fix: VideoCombine auto-detects filename pattern (fixes %06d vs %08d mismatch) 2026-01-22 19:11:37 +08:00
Hawk Lee b3313cdade perf: Maximize I/O with PNG Level 0 (No Compression) 2026-01-22 19:01:38 +08:00
Hawk Lee f51c0f5e5b fix: Set default filename_pattern to %08d for seamless Disk Mode integration 2026-01-22 18:53:01 +08:00
Hawk Lee 5b0a766908 perf: Optimize Disk Mode speed (Parallel I/O + Fast PNG Compression) 2026-01-22 18:51:42 +08:00
Hawk Lee 5e5f7635c0 feat: Add frames_directory input/output to BodySway for OOM-safe disk mode 2026-01-22 18:30:18 +08:00
Hawk Lee f22abda3b1 feat: Add save_to_disk mode for Ditto to prevent OOM on large frame counts 2026-01-22 18:19:30 +08:00
Hawk Lee e6f97e233e perf: Optimize Ditto frame→tensor conversion to avoid memory doubling 2026-01-22 18:13:00 +08:00
Hawk Lee 731779be36 perf: True incremental frame release in VideoCombine by converting tensor to list 2026-01-22 16:21:01 +08:00
Hawk Lee e0569820ed perf: Optimize AIIA_VideoCombine memory with incremental frame release 2026-01-22 15:51:27 +08:00
Hawk Lee 8207326979 perf: Aggressive BodySway memory optimization - smaller batches, explicit del, cuda.empty_cache 2026-01-22 15:40:37 +08:00
Hawk Lee b505b521d7 perf: Optimize BodySway memory usage with pre-allocation and batch processing 2026-01-22 15:10:36 +08:00
Hawk Lee 2dd439b8e0 fix: Remove EMA reset on speech onset for smooth mouth opening transition 2026-01-22 14:24:25 +08:00
Hawk Lee a1ab20588d feat: BodySway v2 - Perlin noise, no Y translation, improved defaults (reduces dizziness) 2026-01-22 13:51:07 +08:00
Hawk Lee cbad609b5e fix: Limit BodySway seed to 32-bit for numpy compatibility 2026-01-22 12:54:44 +08:00
Hawk Lee 1188bd20e1 refactor: Use global valid region for consistent sway range across all frames 2026-01-22 12:50:41 +08:00
Hawk Lee 8c72c6d0c0 feat: Add per-frame runtime boundary check in BodySway to guarantee no black pixels 2026-01-22 12:48:10 +08:00
Hawk Lee 3294f2bbd6 fix: Account for rotation margin in BodySway to prevent black corners 2026-01-22 12:43:44 +08:00
Hawk Lee 09aa490f00 refactor: Simplify BodySway to use single crop_ratio parameter (auto-calc sway amplitude) 2026-01-22 12:41:37 +08:00
Hawk Lee 0306b7abdc feat: Add AIIA_BodySway node for simulating subtle body movement via crop/rotation 2026-01-22 12:29:37 +08:00
Hawk Lee 80317d47ae chore: Update icon to use assets/logo.png, fix README formatting 2026-01-22 09:57:58 +08:00
Hawk Lee 81872d74e7 chore: Bump version to 1.9.10 (Soft Reset + mouth_smoothing docs) 2026-01-22 09:52:23 +08:00
Hawk Lee 9778d2c654 refactor: Implement Soft Reset - only reset seed, let model continue from idle naturally 2026-01-22 09:33:40 +08:00
Hawk Lee d6a4182edd fix: Apply warping at correct frame offset (fuse_r2_s) instead of index 0 2026-01-22 01:37:24 +08:00
Hawk Lee 27800ce3f1 fix: Expand Prediction Warping to ALL kp dimensions (incl. expression) to prevent full reset snap 2026-01-22 01:16:07 +08:00
Hawk Lee eb3227a78b fix: Reset EMA on speech onset to prevent stale closed-mouth bleed 2026-01-22 01:11:43 +08:00
Hawk Lee eff970828f feat: Expose mouth_smoothing parameter (None/Light/Normal/Heavy) 2026-01-22 01:06:44 +08:00
Hawk Lee 50c628f3d6 feat: Add Expression Temporal Smoothing (EMA) to prevent inhumanly fast mouth movements 2026-01-22 01:02:28 +08:00
Hawk Lee 0cea1ba9d6 chore: Bump version to 1.9.3 (fix silence_release NameError) 2026-01-22 00:48:54 +08:00
Hawk Lee 175288f6e9 chore: Bump version to 1.9.2; update description with Ditto performance claims 2026-01-22 00:45:37 +08:00
Hawk Lee ed65157bd0 chore: Group FLOAT with Talking Head models in metadata 2026-01-21 01:16:25 +08:00
Hawk Lee b2453bb325 chore: Update pyproject.toml description and keywords for Ditto/FLOAT 2026-01-21 01:15:14 +08:00
Hawk Lee 030c03a48d chore: bump version to 1.9.0 and update metadata for EchoMimic V3 2026-01-20 11:34:06 +08:00
Hawk Lee 3af5060421 chore: bump version to 1.8.4 for registry release 2026-01-19 13:44:49 +08:00
Hawk Lee ae1864a57b fix: VibeVoice speed control and add sox dependency 2026-01-19 12:51:42 +08:00
Hawk Lee 1ac8d65f01 docs: Update README for VibeVoice optional reference audio (v1.8.3) 2026-01-07 10:49:23 +08:00
Hawk Lee 5ee0a975d2 chore: Bump version to 1.8.2 to trigger registry fix verification 2026-01-06 20:04:37 +08:00
Hawk Lee 6f5446c20e feat: Workflow Caching Optimization (v1.8.1) 
- Modify Script Parser to return clean tts_data (enabling TTS caching)
- Add AIIA_Segment_Merge node to re-attach visual tags post-TTS
- Update docs with Separate Streams workflow
 2026-01-05 15:15:35 +08:00
Hawk Lee 8d933f0205 feat: Interactive Teaching Web Export (v1.8.0) 
- Add AIIA_Web_Export node for generating HTML player
- Update Script Parser to support (Visual: url) tags
- Update Dialogue TTS to propagate visual tags to segments
- Support relative paths for visual content
 2026-01-05 15:07:38 +08:00
Hawk Lee 8f566dc16e fix: Generate per-speaker ASS styles with distinct colors (v1.7.2) 
- Automatically assign colors to speakers in ASS output
- Fixes issue where all speakers used default white style
 2026-01-05 14:01:59 +08:00
Hawk Lee 4c34904cf4 feat: Add AIIA_Subtitle_Preview node (v1.7.1) 
- New node for previewing audio with synchronized subtitles
- Supports SRT and basic ASS styles (Color, Font Size)
- Includes Javascript frontend for timeline playback
 2026-01-05 13:46:15 +08:00
Hawk Lee 57285ec603 feat: Add Subtitle Generation support (v1.7.0) 
- Add AIIA_Subtitle_Gen node for SRT/ASS generation
- Update AIIA_Dialogue_TTS to output precise timestamps (segments_info)
- Implement Smart Batch Interpolation for VibeVoice timestamps
- Update README and Walkthrough
 2026-01-05 13:33:57 +08:00
Hawk Lee e76af6119e fix: resolve submodule error by vendoring CosyVoice lib + bump version to 1.6.2 2026-01-04 23:41:10 +08:00