text
stringlengths
8
1.22k
index: 10
item_id: sounding_order_01
category: sounding_action
label: clap_slapping_the_thigh
seed: 52
modality_combination: Action (sounding)
content: {"action_1": "clap", "action_2": "slapping_the_thigh"}
temporal_constraint: {"type": "order", "value": {"sequence": ["action_1", "action_2"]}}
prompt: A static medium shot shows one person standing in a plain indoor room against an uncluttered background. The person first claps once, then immediately slaps their thigh once, in that exact order. Each clap and thigh slap makes a clear sound at the moment of contact, and there are no other sounds. The camera sta...
command: /home/coder/miniforge3/envs/ltx23/bin/python -m ltx_pipelines.ti2vid_one_stage --checkpoint-path /home/coder/T2V_Gen/LTX-2/checkpoints/ltx-2.3-22b-dev.safetensors --gemma-root /home/coder/T2V_Gen/LTX-2/checkpoints/gemma-3-12b-it --prompt 'A static medium shot shows one person standing in a plain indoor room ag...
index: 11
item_id: sounding_order_02
category: sounding_action
label: clap_slapping_the_thigh
seed: 53
modality_combination: Action (sounding)
content: {"action_1": "clap", "action_2": "slapping_the_thigh"}
temporal_constraint: {"type": "order", "value": {"sequence": ["action_2", "action_1"]}}
prompt: A static medium shot shows one person in a plain indoor room facing the camera. The person first slaps their thigh, then claps their hands once. The audio captures the thigh slap before the clap, and each sound is clear and synchronized with the motion. No other actions occur, no other people enter, no music pl...
command: /home/coder/miniforge3/envs/ltx23/bin/python -m ltx_pipelines.ti2vid_one_stage --checkpoint-path /home/coder/T2V_Gen/LTX-2/checkpoints/ltx-2.3-22b-dev.safetensors --gemma-root /home/coder/T2V_Gen/LTX-2/checkpoints/gemma-3-12b-it --prompt 'A static medium shot shows one person in a plain indoor room facing the ...
index: 12
item_id: sounding_rhythm_01
category: sounding_action
label: clap
seed: 54
modality_combination: Action (sounding)
content: {"action": "clap"}
temporal_constraint: {"type": "rhythm", "value": {"pattern": "steady_continuous"}}
prompt: A static medium shot shows one person standing in a plain indoor room against an uncluttered background. The person claps at a steady rhythm throughout the clip, with each clap producing a clear sound at the moment the hands meet. The camera remains fixed and the framing does not change. There are no extra acti...
command: /home/coder/miniforge3/envs/ltx23/bin/python -m ltx_pipelines.ti2vid_one_stage --checkpoint-path /home/coder/T2V_Gen/LTX-2/checkpoints/ltx-2.3-22b-dev.safetensors --gemma-root /home/coder/T2V_Gen/LTX-2/checkpoints/gemma-3-12b-it --prompt 'A static medium shot shows one person standing in a plain indoor room ag...
index: 13
item_id: sounding_relative_spacing_01
category: sounding_action
label: clap
seed: 55
modality_combination: Action (sounding)
content: {"action": "clap"}
temporal_constraint: {"type": "relative_spacing", "value": {"description": "perform the action twice with a brief pause between the two repetitions"}}
prompt: A static medium shot shows a single person standing in a plain indoor room against an uncluttered background. The person claps exactly twice, with a brief pause between the two claps. Each clap makes a sharp sound at the moment the hands meet. The camera stays still and centered on the subject throughout. There...
command: /home/coder/miniforge3/envs/ltx23/bin/python -m ltx_pipelines.ti2vid_one_stage --checkpoint-path /home/coder/T2V_Gen/LTX-2/checkpoints/ltx-2.3-22b-dev.safetensors --gemma-root /home/coder/T2V_Gen/LTX-2/checkpoints/gemma-3-12b-it --prompt 'A static medium shot shows a single person standing in a plain indoor ro...
index: 14
item_id: sounding_absolute_timing_01
category: sounding_action
label: clap
seed: 56
modality_combination: Action (sounding)
content: {"action": "clap"}
temporal_constraint: {"type": "absolute_timing", "value": {"description": "perform the action once, wait exactly 1.0 second, then perform it again"}}
prompt: A static medium shot shows one person standing in a plain indoor room against a simple empty background. The person claps exactly once, waits exactly 1.0 second, then claps exactly once again. Each clap makes a clear sharp sound at the moment the hands meet, and there are no other actions. The camera stays comp...
command: /home/coder/miniforge3/envs/ltx23/bin/python -m ltx_pipelines.ti2vid_one_stage --checkpoint-path /home/coder/T2V_Gen/LTX-2/checkpoints/ltx-2.3-22b-dev.safetensors --gemma-root /home/coder/T2V_Gen/LTX-2/checkpoints/gemma-3-12b-it --prompt 'A static medium shot shows one person standing in a plain indoor room ag...
index: 15
item_id: sounding_start_stop_01
category: other
label: clap
seed: 57
modality_combination: Speech + Action (sounding)
content: {"speech_1": "go", "speech_2": "stop", "action": "clap"}
temporal_constraint: {"type": "start_stop", "value": {"start_on": "speech_1", "stop_on": "speech_2", "description": "start the action when the person says go and stop the action when the person says stop"}}
prompt: A static medium shot shows one person standing in a plain indoor room facing the camera. The person says “go,” then claps repeatedly, and the clapping continues until the person says “stop,” at which point the clapping ends immediately. Each clap produces a clear sound exactly when the hands meet, and the audio...
command: /home/coder/miniforge3/envs/ltx23/bin/python -m ltx_pipelines.ti2vid_one_stage --checkpoint-path /home/coder/T2V_Gen/LTX-2/checkpoints/ltx-2.3-22b-dev.safetensors --gemma-root /home/coder/T2V_Gen/LTX-2/checkpoints/gemma-3-12b-it --prompt 'A static medium shot shows one person standing in a plain indoor room fa...
index: 16
item_id: speech_count_01
category: speech
label: help
seed: 58
modality_combination: Speech
content: {"speech": "help"}
temporal_constraint: {"type": "count", "value": {"count": 1}}
prompt: A static medium close-up shot shows one person in a plain indoor room facing the camera. The person says “help” exactly once, clearly and at normal speaking volume. The audio contains only that single spoken word and no other speech or sound. Keep the camera still with no extra actions, no other people, no subt...
command: /home/coder/miniforge3/envs/ltx23/bin/python -m ltx_pipelines.ti2vid_one_stage --checkpoint-path /home/coder/T2V_Gen/LTX-2/checkpoints/ltx-2.3-22b-dev.safetensors --gemma-root /home/coder/T2V_Gen/LTX-2/checkpoints/gemma-3-12b-it --prompt 'A static medium close-up shot shows one person in a plain indoor room fa...
index: 17
item_id: speech_count_02
category: speech
label: help
seed: 59
modality_combination: Speech
content: {"speech": "help"}
temporal_constraint: {"type": "count", "value": {"count": 2}}
prompt: A static medium close-up shot shows one person in a plain indoor room facing the camera. The person says “help” exactly twice, with a clear pause between the two spoken words. The audio contains only the two spoken words and no other sounds. The camera stays completely still, and the clip ends immediately after...
command: /home/coder/miniforge3/envs/ltx23/bin/python -m ltx_pipelines.ti2vid_one_stage --checkpoint-path /home/coder/T2V_Gen/LTX-2/checkpoints/ltx-2.3-22b-dev.safetensors --gemma-root /home/coder/T2V_Gen/LTX-2/checkpoints/gemma-3-12b-it --prompt 'A static medium close-up shot shows one person in a plain indoor room fa...
index: 18
item_id: speech_order_01
category: speech
label: speech_order_01
seed: 60
modality_combination: Speech
content: {"speech_1": "help", "speech_2": "no"}
temporal_constraint: {"type": "order", "value": {"sequence": ["speech_1", "speech_2"]}}
prompt: A static medium close-up shot shows one person seated in a plain indoor room, facing the camera. The person speaks clearly and only twice, saying “help” first and then “no” in that order. The audio contains only those two spoken words, with no extra speech or background noise. The camera stays still throughout,...
command: /home/coder/miniforge3/envs/ltx23/bin/python -m ltx_pipelines.ti2vid_one_stage --checkpoint-path /home/coder/T2V_Gen/LTX-2/checkpoints/ltx-2.3-22b-dev.safetensors --gemma-root /home/coder/T2V_Gen/LTX-2/checkpoints/gemma-3-12b-it --prompt 'A static medium close-up shot shows one person seated in a plain indoor ...
index: 19
item_id: speech_order_02
category: speech
label: speech_order_02
seed: 61
modality_combination: Speech
content: {"speech_1": "help", "speech_2": "no"}
temporal_constraint: {"type": "order", "value": {"sequence": ["speech_2", "speech_1"]}}
prompt: A static medium close-up shot shows one person in a plain indoor room, facing the camera. The person says "no" first, then says "help" afterward, with the two words spoken in that exact order and clearly separated. The audio contains only their voice, with no extra words, no background music, and no other sound...
command: /home/coder/miniforge3/envs/ltx23/bin/python -m ltx_pipelines.ti2vid_one_stage --checkpoint-path /home/coder/T2V_Gen/LTX-2/checkpoints/ltx-2.3-22b-dev.safetensors --gemma-root /home/coder/T2V_Gen/LTX-2/checkpoints/gemma-3-12b-it --prompt 'A static medium close-up shot shows one person in a plain indoor room, f...