<h2 id="ltx-2.3-fart-sfx-lora-summary">LTX 2.3 Fart SFX LoRA — Summary</h2>What it is: A Text-to-Audio (T2A) LoRA for LTX 2.3 that adds a specific fart sound effect to generations, invoked via the word "fart." This LoRA works well with LTX 2.3 I2V, I haven't tested it with T2V. It is audio only, so you have to ask for the sound effect by using the trigger word "fart".Training specs:<ul><li>Mode: T2A (audio-only) — targets only <code>audio_attn1</code>, <code>audio_attn2</code>, <code>audio_ff</code> modules, doesn't touch video or speech-related weights</li><li>Rank 16 / alpha 16, trained unquantized (bf16) for 2000 steps</li><li>Dataset: 29 training clips (of 32 total — 1 too short for any bucket, 2 held out for validation) that I generated and hand-captioned with texture descriptions (e.g. "long wet," "small," "walking triple") rather than the literal word "fart," so the trigger word alone carries the concept</li><li>Final loss 0.5122, trained in ~8.4 hours overnight</li><li>Validated against 8 prompts (3 in-distribution, 3 novel out-of-distribution, 2 held-out) </li></ul>

Files with SHA256: e01018c9289b1e675b7543b6d5ad7800c0f3e175f1fc65047076a2bc25c0b63d