$ ./tools/sfxgen --prompt "Soft UI button click: small bronze tick with a tiny wooden knock. Isolated sound effect, no music, no voices, no reverb tail beyond 0.3 s, clean start." --seconds 0.2 --takes 3 --out audio/source/ui/sfx-ui-click.mp3 # cwd /Users/king/Desktop/Workspace/SRO/zoen # exit 0 Generating take 1/3: /Users/king/Desktop/Workspace/SRO/zoen/audio/source/ui/sfx-ui-click.mp3 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ SA3 → MLX text-to-audio ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ prompt 'Soft UI button click: small bronze tick with a tiny wooden knock. Isolated sound effect, no music, no voices, no reverb tail beyond 0.3 s, clean start.' dit sm-sfx decoder same-s σmax 1.00 seconds 1.5s steps 8 seed 459088386 DiT dtype fp16 cfg 1.0 T_lat 17 (1.58s → trimmed to 1.5s) [1/5] T5Gemma encode ································· 627 ms embeds (1, 256, 768) mlx.core.float16 [2/5] Conditioning ··································· 5 ms cross_attn (1, 257, 768) global (1, 768) [3/5] DiT — load + sample (8 steps, σmax=1.00) load 0.5s schedule 1.000 · 0.994 · 0.984 · 0.958 · 0.891 · 0.746 · 0.512 · 0.274 · 0.000 sample 1174 ms (147 ms/step) [4/5] Decoder (same-s, FP32) load 109 ms decode chunked (chunk=8, ovl=2) → 36 ms patches (1, 512, 272) [5/5] Unpatch + write WAV ···························· 3 ms audio (2, 66150) peak 0.337 rms 0.065 Peak RAM by stage T5Gemma encode ███████████············· 708 MB Conditioning █████████··············· 539 MB DiT load ██████████████·········· 878 MB DiT sample ████████████████████████ 1.45 GB ← max Decoder load ███····················· 210 MB Decode ███████················· 438 MB Unpatch + WAV ███····················· 211 MB ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ done 2.52s wall → 1.5s audio → 0.60× realtime peak RAM 1.45 GB seed 459088386 ▸ saved /Users/king/Desktop/Workspace/SRO/zoen/audio/source/ui/.sfxgen-vmbzziz1/source.wav ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ {"path": "/Users/king/Desktop/Workspace/SRO/zoen/audio/source/ui/sfx-ui-click.mp3", "sample_rate": 44100, "channels": 2, "seconds": 0.2, "peak_dbfs": -12.199, "bytes": 5314, "boundary_jump": 0.0005893469788134098} Generating take 2/3: /Users/king/Desktop/Workspace/SRO/zoen/audio/source/ui/sfx-ui-click_2.mp3 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ SA3 → MLX text-to-audio ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ prompt 'Soft UI button click: small bronze tick with a tiny wooden knock. Isolated sound effect, no music, no voices, no reverb tail beyond 0.3 s, clean start.' dit sm-sfx decoder same-s σmax 1.00 seconds 1.5s steps 8 seed 707514594 DiT dtype fp16 cfg 1.0 T_lat 17 (1.58s → trimmed to 1.5s) [1/5] T5Gemma encode ································· 1072 ms embeds (1, 256, 768) mlx.core.float16 [2/5] Conditioning ··································· 33 ms cross_attn (1, 257, 768) global (1, 768) [3/5] DiT — load + sample (8 steps, σmax=1.00) load 0.7s schedule 1.000 · 0.994 · 0.984 · 0.958 · 0.891 · 0.746 · 0.512 · 0.274 · 0.000 sample 1069 ms (134 ms/step) [4/5] Decoder (same-s, FP32) load 85 ms decode chunked (chunk=8, ovl=2) → 40 ms patches (1, 512, 272) [5/5] Unpatch + write WAV ···························· 2 ms audio (2, 66150) peak 0.343 rms 0.066 Peak RAM by stage T5Gemma encode ███████████············· 708 MB Conditioning █████████··············· 539 MB DiT load ██████████████·········· 878 MB DiT sample ████████████████████████ 1.45 GB ← max Decoder load ███····················· 210 MB Decode ███████················· 438 MB Unpatch + WAV ███····················· 211 MB ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ done 3.11s wall → 1.5s audio → 0.48× realtime peak RAM 1.45 GB seed 707514594 ▸ saved /Users/king/Desktop/Workspace/SRO/zoen/audio/source/ui/.sfxgen-8691m371/source.wav ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ {"path": "/Users/king/Desktop/Workspace/SRO/zoen/audio/source/ui/sfx-ui-click_2.mp3", "sample_rate": 44100, "channels": 2, "seconds": 0.2, "peak_dbfs": -11.04, "bytes": 5157, "boundary_jump": 0.0004970845766365528} Generating take 3/3: /Users/king/Desktop/Workspace/SRO/zoen/audio/source/ui/sfx-ui-click_3.mp3 ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ SA3 → MLX text-to-audio ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ prompt 'Soft UI button click: small bronze tick with a tiny wooden knock. Isolated sound effect, no music, no voices, no reverb tail beyond 0.3 s, clean start.' dit sm-sfx decoder same-s σmax 1.00 seconds 1.5s steps 8 seed 751880703 DiT dtype fp16 cfg 1.0 T_lat 17 (1.58s → trimmed to 1.5s) [1/5] T5Gemma encode ································· 355 ms embeds (1, 256, 768) mlx.core.float16 [2/5] Conditioning ··································· 3 ms cross_attn (1, 257, 768) global (1, 768) [3/5] DiT — load + sample (8 steps, σmax=1.00) load 0.3s schedule 1.000 · 0.994 · 0.984 · 0.958 · 0.891 · 0.746 · 0.512 · 0.274 · 0.000 sample 940 ms (118 ms/step) [4/5] Decoder (same-s, FP32) load 101 ms decode chunked (chunk=8, ovl=2) → 43 ms patches (1, 512, 272) [5/5] Unpatch + write WAV ···························· 3 ms audio (2, 66150) peak 0.221 rms 0.056 Peak RAM by stage T5Gemma encode ███████████············· 708 MB Conditioning █████████··············· 539 MB DiT load ██████████████·········· 878 MB DiT sample ████████████████████████ 1.45 GB ← max Decoder load ███····················· 210 MB Decode ███████················· 438 MB Unpatch + WAV ███····················· 211 MB ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ done 1.81s wall → 1.5s audio → 0.83× realtime peak RAM 1.45 GB seed 751880703 ▸ saved /Users/king/Desktop/Workspace/SRO/zoen/audio/source/ui/.sfxgen-yfuojdg5/source.wav ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ {"path": "/Users/king/Desktop/Workspace/SRO/zoen/audio/source/ui/sfx-ui-click_3.mp3", "sample_rate": 44100, "channels": 2, "seconds": 0.2, "peak_dbfs": -15.076, "bytes": 5157, "boundary_jump": 0.0005695982836186886}