{ "audioMethod": "Original audio decoded to mono 16 kHz and transcribed with both faster-whisper base.en and small.en. Word timestamps are approximate; Veo timing also checked in separate audio windows. Voice performance uses creator listening feedback for Seedance and Veo; MiniMax, Grok and Wan remain unreviewed. No sound-effect-quality or phoneme-level lip-sync score.", "criteria": [ { "group": "performance", "key": "story", "label": "Story adherence", "meaning": "The intended story and character intentions survive: a boast, an unnoticed peel, an accidental fall and an embarrassed cover-up. Looking at the peel changes the joke." }, { "group": "performance", "key": "movement", "label": "Natural movement", "meaning": "Foot contact, loss of balance and landing connect into a readable accident. Cartoon exaggeration is welcome; unexplained floating or a deliberate crouch weakens the action." }, { "group": "performance", "key": "acting", "label": "Expressive acting", "meaning": "Faces and gestures convey the banana's pride, surprise and forced confidence, and the strawberry's skepticism and sarcastic amusement." }, { "group": "performance", "key": "shots", "label": "Shot sequence", "meaning": "Distinct opening, fall, banana and strawberry shots arrive in the requested order. The edits let the viewer follow one event." }, { "group": "world", "key": "reference", "label": "Reference likeness", "meaning": "The fruit heads, yellow tracksuit, red overalls and kitchen match the shared image. This checks the starting design, separately from consistency over time." }, { "group": "world", "key": "continuity", "label": "Character consistency", "meaning": "The same faces, fruit shapes, clothing and body designs persist through movement and camera changes, without obvious identity changes or deformation." }, { "group": "world", "key": "world", "label": "Scene & prop continuity", "meaning": "The kitchen, character positions and banana peel stay coherent across shots. Props should not duplicate or change without a story reason." }, { "group": "audio", "key": "speech", "label": "Dialogue accuracy", "meaning": "The original audio contains \"Totally planned.\" followed by \"Sure it was.\" Both phrases are corroborated by two Whisper models. Voice quality is a separate check." }, { "group": "audio", "key": "timing", "label": "Dialogue timing", "meaning": "The first line follows the landing and the reply accompanies the strawberry's reaction. This checks story-beat timing, not precise mouth synchronization or speaker identity." }, { "group": "audio", "key": "voice", "label": "Voice & delivery", "meaning": "Human listening judges whether speech sounds natural and suits the character and comic delivery. The prompt specifies English but leaves voice gender open. Listening pending means no verdict yet." }, { "group": "delivery", "key": "captions", "label": "Subtitle accuracy & timing", "meaning": "Both exact English lines are readable and appear during the corresponding dialogue. Missing, garbled or early text makes the clip harder to publish as delivered." }, { "group": "delivery", "key": "portrait", "label": "Portrait framing", "meaning": "The actual picture fills the requested vertical frame. A portrait file containing a small landscape picture and embedded black bands does not meet this check." } ], "criterionGroups": [ { "key": "performance", "label": "Story & performance" }, { "key": "world", "label": "Characters & world" }, { "key": "audio", "label": "Dialogue & voice" }, { "key": "delivery", "label": "Subtitles & format" } ], "inputs": { "story": "Use the uploaded reference image for both characters and the kitchen. Preserve the banana's yellow tracksuit, the strawberry's red overalls, and both characters' fruit-head designs throughout.\n\nCreate a five-second comedy using the following four shots in order, with three distinct camera cuts. Each cut continues the same event.\n\nShot 1, approximately 0–1 second:\nA medium two-shot of the banana and strawberry. The banana proudly flexes like a bodybuilder, looking at the strawberry with a smug grin. The strawberry watches skeptically. The banana is completely unaware of the loose banana peel on the floor.\n\nShot 2, approximately 1–2 seconds:\nCut to a wide, low-angle shot showing the banana's full body and the peel. Still looking toward the strawberry without noticing the peel, the banana confidently steps forward onto it. The peel slides under its shoe, causing its feet to slip forward. The banana loses balance and lands on its bottom.\n\nShot 3, approximately 2–3.5 seconds:\nCut to a medium close-up of the banana, now seated on the floor. It immediately gives a thumbs-up, forces a proud grin, and says in English, \"Totally planned.\"\n\nShot 4, approximately 3.5–5 seconds:\nCut to a close-up of the strawberry looking down toward the seated banana. The strawberry chuckles, gives a sarcastic smile, and replies in English, \"Sure it was.\"\n\nUse quick slapstick timing and expressive reactions. Synchronize each character's mouth with its own spoken line.\n\nDisplay clear, readable English subtitles at the bottom center of the video. Show \"Totally planned.\" only during the banana's line, then \"Sure it was.\" only during the strawberry's reply. Match the spoken words exactly.\n\nInclude a cartoon slipping sound, a soft landing thud, and the strawberry's chuckle. Keep both spoken lines clearly audible. Maintain consistent character appearances, clothing, kitchen layout, and spatial continuity across all four shots.", "style": "Cinematic stylized 3D fruit characters. Each character's entire head is a whole, uncut fruit, with expressive eyes and a mouth formed directly in its surface. No separate human-skin face, fruit hats, fruit-shaped hair, or fruit helmets. Human body proportions and clothing. Smooth fruit-colored body skin; natural fruit texture only on the head. Not mascots." }, "listeningFeedback": { "recorded_on": "2026-10-10", "source": "First-hand listening feedback in the benchmark creator's page-review message in this conversation.", "method": "Qualitative creator feedback, not a blinded listening panel or an automated voice-quality metric.", "prompt_voice_constraints": { "language": "English", "voice_gender": "unspecified" }, "reviews": [ { "model_id": "seedance-2-mini", "status": "reviewed", "voice_result": "pass", "feedback": "While discussing voice and speech performance, the creator said they were very satisfied with Seedance.", "verbatim_zh": "这里不得不说我自己的直觉来说seedance真是非常满意" }, { "model_id": "veo-3-1-lite", "status": "reviewed", "voice_result": "fail", "feedback": "The creator perceived the strawberry voice as masculine and unnatural. The failure concerns the reported unnatural delivery. Voice gender was not constrained by the frozen prompt.", "verbatim_zh": "veo 的声音草莓变成了男声而且不自然" }, { "model_id": "minimax-h3-max", "status": "listening_pending", "voice_result": "unreviewed" }, { "model_id": "grok-imagine-video", "status": "listening_pending", "voice_result": "unreviewed" }, { "model_id": "wan-2-7", "status": "listening_pending", "voice_result": "unreviewed" } ], "not_established": [ "Voice gender as a prompt-following requirement", "Independent listening verdicts for MiniMax, Grok and Wan", "Speaker assignment or phoneme-level lip synchronization", "Sound-effect quality" ] }, "manifest": { "comparison_kind": "same_story_different_input_modes_and_settings", "default_style": { "file": "research/fruit-video-benchmark-runs/banana-slip-v2/default-style.txt", "source": "packages/media/generation-prompt.ts", "sha256_of_input_utf8": "ab215a8e748b5b0383a22c7db9df411fd62bee94bfa6e21ec2a2cfeace324e22" }, "expected_composed_prompt": { "file": "research/fruit-video-benchmark-runs/banana-slip-v2/expected-composed-prompt.txt", "sha256_of_input_utf8": "f8044921a56b44b2497ee06ea062200edf232a8ff6fe0ce2c6d3e1d8ec253383", "evidence": "Reconstructed from current local Simple composer, not captured from the Production Job.", "production_request_verified": false }, "recorded_date_local": "2026-10-10", "reference": { "file": "apps/web/public/benchmarks/banana-slip-v2/reference.png", "sha256": "1c1310f1bd36b7b67d3a037f9fe089c409a02b49fb3eb623fdd31a69461b4bdd" }, "runs": [ { "bytes": 1033895, "execution_channel": "https://aifruitvideo.com/", "id": "banana-slip-v2-seedance-mini-01", "model_display_name": "Seedance 2.0 Mini", "output_metadata": { "programs": [], "stream_groups": [], "streams": [ { "codec_name": "h264", "codec_type": "video", "width": 480, "height": 854, "r_frame_rate": "24/1" }, { "codec_name": "aac", "codec_type": "audio", "sample_rate": "32000", "channels": 2, "r_frame_rate": "0/0" } ], "format": { "duration": "5.088000" } }, "reference_input_from_user_appshot": { "ui_selected_image_count": 1, "ui_alias": "@Image 1", "input_mode_observed": "generic_reference_image", "first_frame_constraint_verified": false, "upstream_image_bytes_verified": false }, "requested_settings_from_user_appshot": { "duration_seconds": 5, "resolution": "480p", "aspect_ratio": "9:16", "creation_mode": "simple" }, "seed": null, "sha256": "48d7fbc55b65291c11c2d7df0029a93a31ba12687ed0ba1bf141e236a6e51f5d", "video": "/benchmarks/banana-slip-v2/seedance-2-mini-01.mp4" }, { "bytes": 1215288, "execution_channel": "https://aifruitvideo.com/", "id": "banana-slip-v2-veo-lite-01", "model_display_name": "Veo 3.1 Lite", "output_metadata": { "programs": [], "stream_groups": [], "streams": [ { "codec_name": "h264", "codec_type": "video", "width": 720, "height": 1280, "r_frame_rate": "24/1" }, { "codec_name": "aac", "codec_type": "audio", "sample_rate": "48000", "channels": 2, "r_frame_rate": "0/0" } ], "format": { "duration": "4.020000" } }, "reference_input_from_user_appshot": { "input_mode_observed": "first_frame", "ui_label": "First frame", "ui_selected_image_count": 1, "first_frame_constraint_verified": false, "upstream_image_bytes_verified": false }, "requested_settings_from_user_appshot": { "duration_seconds": 4, "resolution": "720p", "aspect_ratio": "9:16", "creation_mode": "simple" }, "seed": null, "sha256": "9285a4623f46e3b5cd37335086b9a10e523e787790d2f7ee427f6aa54d8f68aa", "video": "/benchmarks/banana-slip-v2/veo-3-1-lite-01.mp4" }, { "bytes": 1024196, "execution_channel": "https://aifruitvideo.com/", "id": "banana-slip-v2-grok-imagine-01", "model_display_name": "Grok Imagine Video", "output_metadata": { "programs": [], "stream_groups": [], "streams": [ { "codec_name": "h264", "codec_type": "video", "width": 480, "height": 854, "r_frame_rate": "24/1" }, { "codec_name": "aac", "codec_type": "audio", "sample_rate": "44100", "channels": 2, "r_frame_rate": "0/0" } ], "format": { "duration": "5.041667" } }, "reference_input_from_user_appshot": { "input_mode_observed": "first_frame", "ui_label": "First frame", "ui_selected_image_count": 1, "first_frame_constraint_verified": false, "upstream_image_bytes_verified": false }, "requested_settings_from_user_appshot": { "duration_seconds": 5, "resolution": "480p", "aspect_ratio": "9:16", "creation_mode": "simple" }, "seed": null, "sha256": "a9447f4c4ca53edae04abd478802e101c568f925146606a688c083832bf75f00", "video": "/benchmarks/banana-slip-v2/grok-imagine-video-01.mp4" }, { "bytes": 627202, "execution_channel": "https://aifruitvideo.com/studio", "id": "banana-slip-v2-wan-2-7-01", "model_display_name": "Wan 2.7", "output_metadata": { "programs": [], "stream_groups": [], "streams": [ { "codec_name": "h264", "codec_type": "video", "width": 720, "height": 1280, "r_frame_rate": "30/1" }, { "codec_name": "aac", "codec_type": "audio", "sample_rate": "44100", "channels": 2, "r_frame_rate": "0/0" } ], "format": { "duration": "5.038005" } }, "reference_input_from_user_appshot": { "input_mode_observed": "generic_reference_image", "ui_label": "References", "ui_selected_image_count": 1, "first_frame_constraint_verified": false, "upstream_image_bytes_verified": false }, "requested_settings_from_user_appshot": { "duration_seconds": 5, "resolution": "720p", "aspect_ratio": "9:16", "creation_mode": "studio" }, "seed": null, "sha256": "83e5e39441ee4cb625896ac3a0ef8fa7a65dd9fe0cad008372533764a7664172", "video": "/benchmarks/banana-slip-v2/wan-2-7-01.mp4" }, { "bytes": 910457, "execution_channel": "https://aifruitvideo.com/studio", "id": "banana-slip-v2-minimax-h3-max-01", "model_display_name": "MiniMax H3 Max", "output_metadata": { "programs": [], "stream_groups": [], "streams": [ { "codec_name": "h264", "codec_type": "video", "width": 480, "height": 854, "r_frame_rate": "24/1" }, { "codec_name": "aac", "codec_type": "audio", "sample_rate": "32000", "channels": 2, "r_frame_rate": "0/0" } ], "format": { "duration": "5.184000" } }, "reference_input_from_user_appshot": { "input_mode_observed": "first_frame", "ui_label": "First frame", "input_mode_user_confirmed": true, "mode_evidence": "User explicitly confirmed First frame for this completed clip; current composer also displays First frame.", "completed_clip_image_count": 1, "first_frame_constraint_verified": false, "upstream_image_bytes_verified": false }, "requested_settings_from_user_appshot": { "duration_seconds": 5, "resolution": "480p", "aspect_ratio": "9:16", "creation_mode": "studio" }, "seed": null, "sha256": "815a3eb2440ff99d469263d29352c5df90feb56e43ccb0ee52d05d9943293246", "video": "/benchmarks/banana-slip-v2/minimax-h3-max-01.mp4" } ], "story_prompt": { "file": "research/fruit-video-benchmark-runs/banana-slip-v2/story-prompt.txt", "characters_without_final_newline": 1924, "sha256_of_input_utf8": "35e6f7ba413674164772e86b71a98257f0009a6cfa8eee748bc07d009f24a7bc" }, "study_id": "banana-slip-v2", "timezone": "Asia/Shanghai" }, "reviewedOn": "2026-10-10", "reviews": [ { "audio": "“Totally planned” accompanies the banana’s seated recovery; “Sure it was” accompanies the strawberry’s reply. The creator’s listening feedback was very positive about the voice performance, making this the strongest starting point for the complete spoken joke in this set.", "checks": { "acting": "pass", "captions": "pass", "continuity": "pass", "movement": "pass", "portrait": "pass", "reference": "pass", "shots": "pass", "speech": "pass", "story": "pass", "timing": "pass", "voice": "pass", "world": "partial" }, "file": "seedance-2-mini-01", "id": "seedance-2-mini", "maker": "ByteDance", "mode": "References", "moments": [ { "label": "The slip", "note": "The step leads into a visible loss of balance.", "time": 1.25 }, { "label": "Check the peel", "note": "The landing works, but a second peel-like shape is visible on the floor.", "time": 1.83 }, { "label": "The reply", "note": "The strawberry gets its own reaction shot and subtitle.", "time": 3.5 } ], "name": "Seedance 2.0 Mini", "nextStep": "This is our first setup to try for this scene. Check the floor after the fall: the extra peel-like shape is the main continuity issue to resolve.", "strength": "The banana looks toward the strawberry before the accident. Its surprise, forced grin and thumbs-up sell the cover-up; the strawberry answers with a skeptical, amused expression. The cast stays consistent across all four shots, with readable, well-timed subtitles.", "summary": "The banana’s confidence turns into a believable accident, then an embarrassed attempt to save face. The strawberry’s reaction completes the joke, with separate speaking shots and readable subtitles.", "tag": "Our pick for this story", "verdict": "Our first choice for the complete fruit scene.", "weakness": "An extra peel-like shape appears near the banana after the fall. The faces and outfits remain consistent, but the floor prop changes across the action." }, { "audio": "Both requested lines are present and arrive with the banana’s recovery and strawberry’s reply. The dialogue works in the story sequence; voice naturalness and character fit still await a listening review.", "checks": { "acting": "pass", "captions": "pass", "continuity": "pass", "movement": "partial", "portrait": "pass", "reference": "pass", "shots": "pass", "speech": "pass", "story": "pass", "timing": "pass", "voice": "unreviewed", "world": "pass" }, "file": "minimax-h3-max-01", "id": "minimax-h3-max", "maker": "fal Research · based on MiniMax H3", "mode": "First frame", "moments": [ { "label": "Foot and peel", "note": "The foot lifts near the peel; the contact and sliding cause are unclear.", "time": 1.25 }, { "label": "The cover-up", "note": "A seated thumbs-up, speaking mouth and correct first subtitle carry the joke.", "time": 2.5 }, { "label": "The comeback", "note": "The strawberry reply has its own close-up and matching text.", "time": 4.17 } ], "name": "MiniMax H3 Max", "nextStep": "Check whether the foot-to-peel contact is convincing enough for your joke. The reaction shots and captions work, but the cause of the accident is less clear.", "strength": "The proud opening, seated thumbs-up and amused strawberry reply carry the intended joke. All four shots arrive in order, the faces and outfits stay consistent, and both subtitle lines match the dialogue.", "summary": "The seated thumbs-up, strawberry reaction and matching subtitles make the exchange easy to follow. The fall is less convincing because the shoe does not clearly slip on the peel.", "tag": "Clear joke, weaker peel contact", "verdict": "A strong alternative, with a less convincing slip.", "weakness": "In the wide shot, the foot moves near or over the peel before the banana lifts into the fall. The peel does not clearly slide under the shoe as requested." }, { "audio": "Both English lines are present, but the banana starts speaking around the tumble and recovery, before the requested seated speaking shot. The words are right; their placement weakens the pause before the cover-up. Voice performance still awaits a listening review.", "checks": { "acting": "pass", "captions": "partial", "continuity": "pass", "movement": "pass", "portrait": "pass", "reference": "pass", "shots": "pass", "speech": "pass", "story": "partial", "timing": "partial", "voice": "unreviewed", "world": "pass" }, "file": "grok-imagine-video-01", "id": "grok-imagine-video", "maker": "xAI", "mode": "First frame", "moments": [ { "label": "The setup", "note": "The banana looks down toward the peel, despite the unaware instruction.", "time": 0.83 }, { "label": "The tumble", "note": "The banana falls onto its back before recovering to a seated position.", "time": 2 }, { "label": "Early reply text", "note": "The reply subtitle is already visible while the banana is still in close-up.", "time": 3.25 } ], "name": "Grok Imagine Video", "nextStep": "Retime the dialogue and captions for a cleaner exchange. Keeping the original accidental setup would require a new take: the banana needs to miss the peel, rather than look at it.", "strength": "The broad backward tumble is readable as cartoon slapstick. A proud recovery and an amused strawberry close-up finish the scene. Character designs and outfits hold together across the four shots.", "summary": "The exaggerated backward tumble gives the scene a big physical reaction. But looking down at the peel before stepping changes the joke: the banana appears aware of the hazard instead of accidentally losing its confidence.", "tag": "Big tumble, changed setup", "verdict": "Expressive slapstick that misses part of the story instruction.", "weakness": "“Sure it was.” appears during the banana close-up, before the strawberry speaking shot. The first subtitle also lingers after the first spoken phrase, so the text handoff needs attention." }, { "audio": "The banana’s line starts during the opening pose, and the reply arrives before the strawberry’s close-up. The creator also found the strawberry’s voice unnatural and masculine-sounding. Voice gender was unspecified, so the voice verdict concerns the unnatural delivery.", "checks": { "acting": "partial", "captions": "fail", "continuity": "pass", "movement": "fail", "portrait": "pass", "reference": "pass", "shots": "fail", "speech": "pass", "story": "partial", "timing": "fail", "voice": "fail", "world": "pass" }, "file": "veo-3-1-lite-01", "id": "veo-3-1-lite", "maker": "Google", "mode": "First frame", "moments": [ { "label": "Early dialogue", "note": "The audio already contains the first line during the opening pose.", "time": 0 }, { "label": "The descent", "note": "The banana crouches toward the floor rather than making a clear slipping fall.", "time": 1.08 }, { "label": "Subtitle check", "note": "The visible generated text does not reproduce the requested English.", "time": 2.17 } ], "name": "Veo 3.1 Lite", "nextStep": "Try again with story timing that fits the selected duration. Check the accidental fall, voice performance and exact subtitle text before keeping the result.", "strength": "The reference fruit heads, outfits and kitchen remain recognizable, and the picture fills the portrait frame. The faces animate clearly, although the broad smiles convey the embarrassed cover-up and sarcasm less well.", "summary": "The characters look familiar, but the banana seems to crouch into a seated pose. Early dialogue, an unnatural voice performance and incorrect subtitles interrupt the intended comedy.", "tag": "Movement and voice need work", "verdict": "The cast looks right; the action and spoken performance need work.", "weakness": "The descent looks like a deliberate crouch, so the accidental slip never reads naturally. Only three main shots appear, and the generated subtitles do not match the requested lines. This run was also shorter than the story timing requested." }, { "audio": "Both English lines are present and broadly align with the seated banana and strawberry reaction. The spoken exchange follows the ending; voice naturalness and character fit still await a listening review.", "checks": { "acting": "pass", "captions": "fail", "continuity": "pass", "movement": "pass", "portrait": "fail", "reference": "pass", "shots": "pass", "speech": "pass", "story": "partial", "timing": "pass", "voice": "unreviewed", "world": "pass" }, "file": "wan-2-7-01", "id": "wan-2-7", "maker": "Alibaba", "mode": "References", "moments": [ { "label": "The full frame", "note": "The black bands belong to the MP4 itself; the player has not added them.", "time": 0 }, { "label": "The fall", "note": "The slip and airborne fall read clearly inside the wide picture.", "time": 1.25 }, { "label": "Missing subtitles", "note": "The strawberry reaction has no visible English subtitle.", "time": 4.25 } ], "name": "Wan 2.7", "nextStep": "Resolve the picture’s embedded black bands and add captions before publishing a vertical version. Check the resulting crop so both characters remain readable on a phone.", "strength": "The loss of balance and landing read as an accident. The banana's forced grin and thumbs-up, followed by the strawberry's amused reaction, carry the ending. Faces and clothing stay recognizable through four shots.", "summary": "The fall and embarrassed recovery carry the joke, but a landscape scene sits inside the vertical file. The black bands make the characters much smaller on a phone, and the requested subtitles are missing.", "tag": "Needs vertical framing and captions", "verdict": "A readable joke that needs work before a vertical release.", "weakness": "The opening skips the requested flexing pose. More visibly, a landscape picture sits between large embedded black bands despite the portrait setting, and the requested English subtitles are absent." } ], "scope": "One original output per model; same story and reference, different input modes and settings. No model-wide success rates.", "title": "AI Fruit Video Models Compared: The Banana-Peel Test", "transcripts": { "method": { "source": "Original MP4 AAC audio, decoded by ffmpeg to mono 16 kHz WAV", "engine": "faster-whisper", "models": [ "base.en", "small.en" ], "language": "en", "condition": "No reference transcript or initial prompt was supplied. Word timestamps are approximate. Both recognizers share the Whisper architecture; agreement is corroboration, not a human listening review.", "unscored": [ "vocal performance", "sound-effect quality", "speaker identity", "fine phoneme-to-mouth synchronization" ] }, "base_en": { "grok-imagine-video-01": [ { "start": 1.7200000000000002, "end": 4.5, "text": " Totally planned. Sure it was.", "no_speech_prob": 0.09914341568946838, "words": [ { "start": 1.7200000000000002, "end": 2.52, "word": " Totally", "probability": 0.5264379382133484 }, { "start": 2.52, "end": 2.92, "word": " planned.", "probability": 0.9926940202713012 }, { "start": 3.66, "end": 3.94, "word": " Sure", "probability": 0.921741783618927 }, { "start": 3.94, "end": 4.22, "word": " it", "probability": 0.9511988162994384 }, { "start": 4.22, "end": 4.5, "word": " was.", "probability": 0.9985815286636353 } ] } ], "minimax-h3-max-01": [ { "start": 1.82, "end": 4.76, "text": " totally planned sure it was", "no_speech_prob": 0.047198645770549774, "words": [ { "start": 1.82, "end": 2.78, "word": " totally", "probability": 0.06780370324850082 }, { "start": 2.78, "end": 3.26, "word": " planned", "probability": 0.976383090019226 }, { "start": 3.26, "end": 4.24, "word": " sure", "probability": 0.5266003608703613 }, { "start": 4.24, "end": 4.52, "word": " it", "probability": 0.9716330170631408 }, { "start": 4.52, "end": 4.76, "word": " was", "probability": 0.9982524514198304 } ] } ], "seedance-2-mini-01": [ { "start": 1.919999999999999, "end": 4.74, "text": " totally planned sure it was", "no_speech_prob": 0.04574478045105934, "words": [ { "start": 1.919999999999999, "end": 2.8, "word": " totally", "probability": 0.13650646805763245 }, { "start": 2.8, "end": 3.24, "word": " planned", "probability": 0.9865579009056092 }, { "start": 3.24, "end": 4.26, "word": " sure", "probability": 0.7230377197265625 }, { "start": 4.26, "end": 4.56, "word": " it", "probability": 0.9757964611053468 }, { "start": 4.56, "end": 4.74, "word": " was", "probability": 0.997905969619751 } ] } ], "veo-3-1-lite-01": [ { "start": 0, "end": 2.58, "text": " Totally planned sure it was", "no_speech_prob": 0.08624692261219025, "words": [ { "start": 0, "end": 0.46, "word": " Totally", "probability": 0.5830896496772766 }, { "start": 0.46, "end": 0.8, "word": " planned", "probability": 0.9905797839164734 }, { "start": 0.8, "end": 2.06, "word": " sure", "probability": 0.3437833786010742 }, { "start": 2.06, "end": 2.3, "word": " it", "probability": 0.95634925365448 }, { "start": 2.3, "end": 2.58, "word": " was", "probability": 0.9986045956611632 } ] } ], "wan-2-7-01": [ { "start": 2.379999999999999, "end": 4.84, "text": " Totally planned, sure it was.", "no_speech_prob": 0.11744801700115204, "words": [ { "start": 2.379999999999999, "end": 2.98, "word": " Totally", "probability": 0.6705289483070374 }, { "start": 2.98, "end": 3.4, "word": " planned,", "probability": 0.9941245913505554 }, { "start": 3.7, "end": 4.44, "word": " sure", "probability": 0.9080293774604796 }, { "start": 4.44, "end": 4.64, "word": " it", "probability": 0.99473237991333 }, { "start": 4.64, "end": 4.84, "word": " was.", "probability": 0.9990633130073548 } ] } ] }, "small_en": { "grok-imagine-video-01": [ { "start": 1.87, "end": 4.61, "text": " Totally planned, sure it was.", "words": [ { "start": 1.87, "end": 2.51, "word": " Totally", "probability": 0.7776481509208679 }, { "start": 2.51, "end": 3.05, "word": " planned,", "probability": 0.9888096451759338 }, { "start": 3.51, "end": 4.07, "word": " sure", "probability": 0.9526873230934144 }, { "start": 4.07, "end": 4.27, "word": " it", "probability": 0.9712343215942384 }, { "start": 4.27, "end": 4.61, "word": " was.", "probability": 0.999474823474884 } ] } ], "minimax-h3-max-01": [ { "start": 2.29, "end": 3.35, "text": " Totally planned.", "words": [ { "start": 2.29, "end": 2.83, "word": " Totally", "probability": 0.8177049160003662 }, { "start": 2.83, "end": 3.35, "word": " planned.", "probability": 0.950693130493164 } ] }, { "start": 3.65, "end": 4.89, "text": " Heh, sure it was.", "words": [ { "start": 3.65, "end": 3.85, "word": " Heh,", "probability": 0.4404875971376896 }, { "start": 4.09, "end": 4.41, "word": " sure", "probability": 0.988653004169464 }, { "start": 4.41, "end": 4.59, "word": " it", "probability": 0.982621133327484 }, { "start": 4.59, "end": 4.89, "word": " was.", "probability": 0.9995877146720886 } ] } ], "seedance-2-mini-01": [ { "start": 1.94, "end": 3.26, "text": " Totally planned!", "words": [ { "start": 1.94, "end": 2.78, "word": " Totally", "probability": 0.688609778881073 }, { "start": 2.78, "end": 3.26, "word": " planned!", "probability": 0.8807564377784729 } ] }, { "start": 4, "end": 4.88, "text": " Sure it was.", "words": [ { "start": 4, "end": 4.42, "word": " Sure", "probability": 0.8667367100715637 }, { "start": 4.42, "end": 4.6, "word": " it", "probability": 0.7837218642234802 }, { "start": 4.6, "end": 4.88, "word": " was.", "probability": 0.9987202882766724 } ] } ], "veo-3-1-lite-01": [ { "start": 0, "end": 2.82, "text": " Totally planned sure it was", "words": [ { "start": 0, "end": 0.46, "word": " Totally", "probability": 0.8582171201705933 }, { "start": 0.46, "end": 1.08, "word": " planned", "probability": 0.9513055682182312 }, { "start": 1.08, "end": 2.12, "word": " sure", "probability": 0.3234717547893524 }, { "start": 2.12, "end": 2.32, "word": " it", "probability": 0.8155879378318787 }, { "start": 2.32, "end": 2.82, "word": " was", "probability": 0.9985792636871338 } ] } ], "wan-2-7-01": [ { "start": 2.48, "end": 3.52, "text": " Totally planned.", "words": [ { "start": 2.48, "end": 2.98, "word": " Totally", "probability": 0.7492742538452148 }, { "start": 2.98, "end": 3.52, "word": " planned.", "probability": 0.9561919569969176 } ] }, { "start": 3.92, "end": 4.92, "text": " Hehe, sure it was.", "words": [ { "start": 3.92, "end": 4.08, "word": " Hehe,", "probability": 0.27224671095609665 }, { "start": 4.28, "end": 4.54, "word": " sure", "probability": 0.9875641465187072 }, { "start": 4.54, "end": 4.7, "word": " it", "probability": 0.9840938448905944 }, { "start": 4.7, "end": 4.92, "word": " was.", "probability": 0.999338924884796 } ] } ] }, "veo_separate_audio_windows": { "vad": [ { "start": 0, "end": 25856 }, { "start": 25856, "end": 53248 }, { "start": 53248, "end": 64512 } ], "0-1.3": [ { "text": " Totally planned", "no_speech": 0.026505015790462497 } ], "1.3-2.8": [ { "text": " Sure it was.", "no_speech": 0.20443062484264377 } ], "2.8-4.02": [ { "text": " Hahaha", "no_speech": 0.19632086157798767 } ] } } }