ComfyUI Node
🧵 AIIA Podcast Stitcher
A ComfyUI node in AIIA/Podcast with 10 inputs and 2 outputs.
🧵 AIIA Podcast Stitcher
- audio_A
- audio_B
- asr_A
- asr_B
- audio
- segments_info
◄split_map—►
◄gap_duration0.25►
◄padding0.10►
◄fade_ms30►
◄use_vadfalse►
◄use_forced_alignfalse►
CategoryAIIA/Podcast
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| split_map | STRING | — | |
| audio_A | AUDIO | — | |
| audio_B | AUDIO | — | |
| asr_A | ASR_RESULT | — | |
| asr_B | ASR_RESULT | — | |
| gap_durationopt | FLOAT | 0.250–2 | 说话人交替时插入的过渡时长（秒） |
| paddingopt | FLOAT | 0.100–0.5 | 每个切片前后保留的呼吸/尾音余量（秒） |
| fade_msopt | INT | 305–100 | 切片首尾的余弦淡入淡出时长（毫秒），越长越平滑 |
| use_vadopt | BOOLEAN | false | 启用 Silero VAD 模型精确检测语音边界（首次使用自动下载 ~2MB 模型） |
| use_forced_alignopt | BOOLEAN | false | 启用 MMS Forced Alignment 字级强制对齐（需要 ~1.2GB 模型，精度最高） |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| audio | AUDIO | — |
| segments_info | STRING | — |