Skip to content

armgit / hypit

Public Imported from GitHub
0

Clone any viral video with AI agents. Not just a script, the whole workflow: swap the face, the words, the B-roll, ship 100 variants in one command, and get your 100M views.

refactor(seedance-kits): author street interview shots explicitly

kashorin181a9743245c
diff --git a/.agents/skills/hypit/references/playbooks/craft/seedance-directing.md b/.agents/skills/hypit/references/playbooks/craft/seedance-directing.md
index dd998d2c..13f0a393 100644
--- a/.agents/skills/hypit/references/playbooks/craft/seedance-directing.md
+++ b/.agents/skills/hypit/references/playbooks/craft/seedance-directing.md
@@ -35,7 +35,7 @@ Choose the Kit by meaning:
| silent B-roll | `broll-v1` | one or more authored images |
| two-person podcast | `podcast-v1` | images 1/2 = final A/B views; audio 1/2 = A/B voices |
| video call | `call-v1` | images 1/2 = reversed call layouts; audio 1/2 = A/B voices |
-| street interview | `street-interview-v1` | image 1 = complete scene; audio 1/2 = interviewer/guest |
+| street interview | `street-interview-v1` | images 1/2/3 = interviewer/guest/shared views; audio 1/2 = interviewer/guest |
| body-motion transfer | `motion-reference-v1` | image 1 = subject; video 1 = motion reference |
| camera-language transfer | `camera-reference-v1` | image 1 = subject; video 1 = camera reference |
diff --git a/.agents/skills/hypit/references/playbooks/formats/street-interview.md b/.agents/skills/hypit/references/playbooks/formats/street-interview.md
index 4c03e8de..e227a305 100644
--- a/.agents/skills/hypit/references/playbooks/formats/street-interview.md
+++ b/.agents/skills/hypit/references/playbooks/formats/street-interview.md
@@ -1,19 +1,20 @@
# Street-interview format
-Use one complete two-person street scene, two explicit Script Roles, and a microphone handoff to
-make the active speaker readable without changing the location or camera topology.
+Use explicit interviewer, guest and shared street views, two Script Roles, and a microphone handoff.
+Write camera changes and performance together in each take's action, in the order they occur.
## Author the SVML program
1. Write Script turns with explicit interviewer and guest Role Cues. Render the Kit dialogue so the
interviewer is `A:` and the guest is `B:`.
-2. Prepare one accepted final scene image containing both people, their full spatial relationship,
- the microphone, wardrobe, street context, camera height, and lens feel.
+2. Prepare three accepted final views in order: image 1 for interviewer A, image 2 for guest B, and
+ image 3 for their shared scene and full spatial relationship.
3. Prepare two clean voice references in the same order: audio 1 for A, audio 2 for B.
-4. Vendor `street-interview-v1.svs`. Select `framing`, `edit-language`, `pacing`, `performance`,
- `reaction`, and `gesture` in an SVS Recipe.
-5. For each Script Segment, use `copy:Render` with dialogue plus an optional English action slot,
- then call `seedance:ReferenceVideo generate-audio="true"` with the scene image and both voices.
+4. Vendor `street-interview-v1.svs`. Select `framing`, `pacing`, `performance`, `reaction`, and
+ `gesture` in an SVS Recipe.
+5. For each Script Segment, use `copy:Render` with dialogue plus an English action that states the
+ camera and performance sequence directly, then call `seedance:ReferenceVideo
+ generate-audio="true"` with all three views and both voices.
6. Normalize each accepted Segment take, produce its `whisperx:SemanticTake`, assemble the results
with `speech:Track`, then add Caption, Media, Typography, and Audio Tracks as needed.
7. Assemble peer Tracks in `film:Film`, render with `render:Video`, and use separate `.svrun`
@@ -21,8 +22,9 @@ make the active speaker readable without changing the location or camera topolog
## Preserve the street-interview grammar
-- Keep one locked or softly handheld two-person perspective. Small operator drift is natural; a new
- camera side, reconstructed street, or unrelated reaction angle is not.
+- Select the shared, interviewer or guest reference directly in the action whenever the camera
+ changes. Preserve the authored camera side, background sector and spatial relationship of the
+ selected view.
- Only the active Role moves their mouth. The listener remains alive through eye focus, breath,
posture, nods, and an appropriate silent reaction.
- A always holds the same microphone. On A lines it remains near A; on B lines A extends it toward B
@@ -43,10 +45,6 @@ make the active speaker readable without changing the location or camera topolog
- Keep exact signs, product labels, and device UI as supplied physical media. Do not ask Seedance to
add editorial cards or captions.
-The primary shared street shot does not need a reverse angle. If a device-facing or proof angle is
-added, its opposing camera position must show a different background sector and landmark set from
-the person-facing image.
-
## Review and reuse
Review speaker/voice assignment, lip-sync, microphone position, listener silence, identities,
diff --git a/.agents/skills/hypit/references/seedance-kits.md b/.agents/skills/hypit/references/seedance-kits.md
index 42ea9ce6..edd035d3 100644
--- a/.agents/skills/hypit/references/seedance-kits.md
+++ b/.agents/skills/hypit/references/seedance-kits.md
@@ -11,7 +11,7 @@ slots carry shot-specific dialogue or action.
| Silent B-roll | `broll-v1.svs` | `story` | one or more authored images |
| Two-person podcast | `podcast-v1.svs` | `dialogue`; optional `action` | images 1/2 = A/B views, audio 1/2 = A/B voices |
| Video call | `call-v1.svs` | `dialogue`; optional `action` | images 1/2 = reversed call layouts, audio 1/2 = A/B voices |
-| Street interview | `street-interview-v1.svs` | `dialogue`; optional `action` | image 1 = complete scene, audio 1/2 = interviewer/guest |
+| Street interview | `street-interview-v1.svs` | `dialogue`; optional `action` | images 1/2/3 = interviewer/guest/shared views, audio 1/2 = interviewer/guest |
| Motion transfer | `motion-reference-v1.svs` | optional `direction` | image 1 = subject, video 1 = motion reference |
| Camera transfer | `camera-reference-v1.svs` | optional `direction` | image 1 = subject, video 1 = camera reference |
@@ -32,7 +32,7 @@ Select these Recipe axes instead of rewriting their prompt blocks:
| `speaker-v1` | `composition-stability`, `camera-motion`, `edit-rhythm`, `performance`, `gesture` |
| `broll-v1` | `material-mode`, `story-shape`, `edit-language`, `camera-language`, `motion-intensity` |
| `podcast-v1` / `call-v1` | `framing`, `edit-language`, `pacing`, `performance`, `reaction`, `gesture` |
-| `street-interview-v1` | `framing`, `edit-language`, `pacing`, `performance`, `reaction`, `gesture` |
+| `street-interview-v1` | `framing`, `pacing`, `performance`, `reaction`, `gesture` |
Use a freeform English prompt only when none of the seven Kit contracts matches. Keep dialogue in
its authored language when it must be spoken verbatim.
diff --git a/docs/quickstart/generation.md b/docs/quickstart/generation.md
index 9b51b895..7a88f15c 100644
--- a/docs/quickstart/generation.md
+++ b/docs/quickstart/generation.md
@@ -253,7 +253,7 @@ Choose the Kit by format, then provide its declared dynamic slots and ordered re
| Silent B-roll | `broll-v1` | `story` | one or more authored images |
| Two-person podcast | `podcast-v1` | `dialogue`; optional `action` | images 1/2 = A/B views; audio 1/2 = A/B voices |
| Video call | `call-v1` | `dialogue`; optional `action` | images 1/2 = reversed call layouts; audio 1/2 = A/B voices |
-| Street interview | `street-interview-v1` | `dialogue`; optional `action` | image 1 = complete scene; audio 1/2 = interviewer/guest |
+| Street interview | `street-interview-v1` | `dialogue`; optional `action` | images 1/2/3 = interviewer/guest/shared views; audio 1/2 = interviewer/guest |
| Motion transfer | `motion-reference-v1` | optional `direction` | image 1 = subject; video 1 = motion reference |
| Camera transfer | `camera-reference-v1` | optional `direction` | image 1 = subject; video 1 = camera reference |
@@ -262,9 +262,9 @@ order.
## Street-interview prompt assembly
-Use `street-interview-v1` for the stable scene, role, microphone, voice and no-overlay contracts.
-Select framing, edit, pacing, performance, reaction and gesture through an SVS Recipe; provide only
-the authored dialogue and optional per-take action as dynamic Text edges:
+Use `street-interview-v1` for the stable view order, role, microphone, voice and no-overlay contracts.
+Select framing, pacing, performance, reaction and gesture through an SVS Recipe. Put each take's
+camera changes and performance in its authored action, in the exact order they should happen:
```svml
<import as="text" from="@hypit/text@1"/>
@@ -272,7 +272,8 @@ the authored dialogue and optional per-take action as dynamic Text edges:
<import as="interview-kit" source="./kits/street-interview-v1.svs"/>
<text:Value id="interview-action">
- Let the guest pause briefly before the answer; keep the microphone handoff natural.
+ Begin with the shared view from @image3 while A asks the question.
+ Cut to B's view from @image2 as B pauses briefly, then answers.
</text:Value>
<text:Render id="interview-prompt"
@@ -285,7 +286,9 @@ the authored dialogue and optional per-take action as dynamic Text edges:
<seedance:ReferenceVideo id="interview-take" model="mini"
prompt={interview-prompt} duration={interview-duration.duration}
resolution="720p" aspect-ratio="9:16" generate-audio="true">
- <seedance:Reference image={interview-scene}/>
+ <seedance:Reference image={interviewer-view}/>
+ <seedance:Reference image={guest-view}/>
+ <seedance:Reference image={shared-view}/>
<seedance:Reference audio={interviewer-voice}/>
<seedance:Reference audio={guest-voice}/>
</seedance:ReferenceVideo>
diff --git a/docs/quickstart/styles.md b/docs/quickstart/styles.md
index dcef6b29..c22317aa 100644
--- a/docs/quickstart/styles.md
+++ b/docs/quickstart/styles.md
@@ -326,7 +326,6 @@ prompt prose. For example:
```svs
interview.street {
framing: soft-handheld;
- edit-language: pause-trim;
pacing: compact;
performance: natural-street;
reaction: active;
@@ -334,7 +333,8 @@ interview.street {
}
```
-`street-interview-v1` reads those six axes. `podcast-v1` and `call-v1` read the corresponding
+`street-interview-v1` reads those five axes. Its per-take `action` carries camera changes in authored
+order. `podcast-v1` and `call-v1` read the corresponding
`framing`, `edit-language`, `pacing`, `performance`, `reaction` and `gesture` axes with their own
finite values. The selected Kit file is the authority for allowed values and defaults.
diff --git a/docs/zh/quickstart/generation.md b/docs/zh/quickstart/generation.md
index 1546211b..63eb4581 100644
--- a/docs/zh/quickstart/generation.md
+++ b/docs/zh/quickstart/generation.md
@@ -221,7 +221,7 @@ Endpoint,不会改变作者图。
| 无声 B-roll | `broll-v1` | `story` | 一张或多张作者声明图片 |
| 双人 Podcast | `podcast-v1` | `dialogue`;可选 `action` | image 1/2 = A/B 视角;audio 1/2 = A/B 声音 |
| 视频通话 | `call-v1` | `dialogue`;可选 `action` | image 1/2 = 相反通话布局;audio 1/2 = A/B 声音 |
-| 街访 | `street-interview-v1` | `dialogue`;可选 `action` | image 1 = 完整场景;audio 1/2 = 采访者/受访者 |
+| 街访 | `street-interview-v1` | `dialogue`;可选 `action` | image 1/2/3 = 采访者/受访者/双人视角;audio 1/2 = 采访者/受访者 |
| 动作迁移 | `motion-reference-v1` | 可选 `direction` | image 1 = 主体;video 1 = 动作参考 |
| 运镜迁移 | `camera-reference-v1` | 可选 `direction` | image 1 = 主体;video 1 = 运镜参考 |
@@ -229,7 +229,7 @@ Endpoint,不会改变作者图。
## 街访 Prompt 组装
-使用 `street-interview-v1` 复用场景、角色、麦克风、音色和无叠加文字契约。构图、剪辑、节奏、表演、反应与手势由 SVS Recipe 选择;动态输入只保留对白与可选的单段动作:
+使用 `street-interview-v1` 复用视角顺序、角色、麦克风、音色和无叠加文字契约。构图、节奏、表演、反应与手势由 SVS Recipe 选择;每段的镜头变化和表演按实际发生顺序直接写在 `action` 中:
```svml
<import as="text" from="@hypit/text@1"/>
@@ -237,7 +237,8 @@ Endpoint,不会改变作者图。
<import as="interview-kit" source="./kits/street-interview-v1.svs"/>
<text:Value id="interview-action">
- Let the guest pause briefly before the answer; keep the microphone handoff natural.
+ Begin with the shared view from @image3 while A asks the question.
+ Cut to B's view from @image2 as B pauses briefly, then answers.
</text:Value>
<text:Render id="interview-prompt"
@@ -250,7 +251,9 @@ Endpoint,不会改变作者图。
<seedance:ReferenceVideo id="interview-take" model="mini"
prompt={interview-prompt} duration={interview-duration.duration}
resolution="720p" aspect-ratio="9:16" generate-audio="true">
- <seedance:Reference image={interview-scene}/>
+ <seedance:Reference image={interviewer-view}/>
+ <seedance:Reference image={guest-view}/>
+ <seedance:Reference image={shared-view}/>
<seedance:Reference audio={interviewer-voice}/>
<seedance:Reference audio={guest-voice}/>
</seedance:ReferenceVideo>
diff --git a/docs/zh/quickstart/styles.md b/docs/zh/quickstart/styles.md
index efa368c7..5457b941 100644
--- a/docs/zh/quickstart/styles.md
+++ b/docs/zh/quickstart/styles.md
@@ -313,7 +313,6 @@ Street Interview、Podcast 与 Call 使用同一套 Recipe 机制,不需要手
```svs
interview.street {
framing: soft-handheld;
- edit-language: pause-trim;
pacing: compact;
performance: natural-street;
reaction: active;
@@ -321,7 +320,7 @@ interview.street {
}
```
-`street-interview-v1` 读取这六个轴。`podcast-v1` 与 `call-v1` 读取同名的 `framing`、
+`street-interview-v1` 读取这五个轴,每段的镜头变化按作者顺序直接写在 `action` 中。`podcast-v1` 与 `call-v1` 读取同名的 `framing`、
`edit-language`、`pacing`、`performance`、`reaction` 和 `gesture` 轴,但使用各自的有限值。允许值和默认值以所选 Kit 文件为准。
模型、分辨率、时长和参考媒体不是模板策略;它们继续存在于精确模型 Surface 与显式图边中。
diff --git a/packages/seedance-kits/README.md b/packages/seedance-kits/README.md
index 4bf4b881..d1900d2d 100644
--- a/packages/seedance-kits/README.md
+++ b/packages/seedance-kits/README.md
@@ -54,7 +54,7 @@ The seven templates are:
- `broll-v1`: silent visual micro-story;
- `podcast-v1`: two fixed podcast views with two voices;
- `call-v1`: two live video-call reverse views;
-- `street-interview-v1`: one shared street scene, two voices and microphone handoff;
+- `street-interview-v1`: interviewer, guest and shared street views, two voices and microphone handoff;
- `motion-reference-v1`: preserve the subject and transfer body motion only;
- `camera-reference-v1`: preserve the subject and transfer camera language only.
diff --git a/packages/seedance-kits/kits/street-interview-v1.svs b/packages/seedance-kits/kits/street-interview-v1.svs
index 5247f632..c89c2d5d 100644
--- a/packages/seedance-kits/kits/street-interview-v1.svs
+++ b/packages/seedance-kits/kits/street-interview-v1.svs
@@ -4,27 +4,21 @@
text-template.street-interview-v1 {
separator: paragraph;
default-framing: soft-handheld;
- default-edit-language: pause-trim;
default-pacing: compact;
default-performance: natural-street;
default-reaction: active;
default-gesture: natural;
}
- text-template.street-interview-v1.block.base { kind: fixed; order: 10; text: "Create a realistic vertical street-interview clip from one final street-interview reference image."; }
- text-template.street-interview-v1.block.reference { kind: fixed; order: 20; text: "REFERENCE CONTRACT: @image1 is the complete scene. Preserve both identities, outfits, street location, props, microphone, lighting, camera height, lens feel, and composition."; }
+ text-template.street-interview-v1.block.base { kind: fixed; order: 10; text: "Create a realistic vertical street-interview clip from three final reference views."; }
+ text-template.street-interview-v1.block.reference { kind: fixed; order: 20; text: "REFERENCE CONTRACT: @image1 is interviewer A's final camera setup, @image2 is guest B's final camera setup, and @image3 is their final shared two-person camera setup. These three supplied setups are a closed shot set, not loose subject references: every authored shot selects exactly one of them as its complete frame, and cuts go directly from one supplied setup to another. Never invent a fourth camera position or pass through an intermediate framing. Within each selected setup, obey the FRAMING direction below. Preserve both identities, outfits, street location, props, microphone, lighting, camera sides, and spatial relationship."; }
text-template.street-interview-v1.block.roles { kind: fixed; order: 30; text: "ROLE MAP: A is the interviewer holding the microphone. B is the guest. Follow A: and B: exactly. A uses @audio1; B uses @audio2; only the current speaker moves their mouth."; }
text-template.street-interview-v1.block.microphone { kind: fixed; order: 40; text: "MICROPHONE CONTRACT: A always holds the same microphone. On A: lines it stays near A's mouth; on B: lines A naturally extends it toward B with realistic wrist, hand, and arm adjustments."; }
- text-template.street-interview-v1.block.scene { kind: fixed; order: 50; text: "SINGLE-SCENE CONTRACT: keep both people in the same street-interview spatial relationship and camera perspective throughout."; }
+ text-template.street-interview-v1.block.order { kind: fixed; order: 50; text: "Follow the authored ACTION DIRECTION in order for camera changes, performance, and physical interaction."; }
text-template.street-interview-v1.block.framing { kind: axis; order: 60; parameter: framing; }
- text-template.street-interview-v1.choice.framing.locked-scene { text: "FRAMING: preserve the reference composition, crop, scale, camera height, and lens feel."; }
- text-template.street-interview-v1.choice.framing.soft-handheld { text: "FRAMING: realistic handheld drift, tiny sway, and operator breathing while preserving composition."; }
- text-template.street-interview-v1.choice.framing.subject-push { text: "FRAMING: a modest handheld push toward the active speaker, then settle back."; }
-
- text-template.street-interview-v1.block.edit { kind: axis; order: 70; parameter: edit-language; }
- text-template.street-interview-v1.choice.edit.continuous { text: "EDITING: a continuous take with real-time reactions, breathing, and conversational timing."; }
- text-template.street-interview-v1.choice.edit.pause-trim { text: "EDITING: trim silent gaps while preserving the same continuous street scene."; }
- text-template.street-interview-v1.choice.edit.beat-cut { text: "EDITING: question, answer, reaction, and punchline land as short beats inside the same scene."; }
+ text-template.street-interview-v1.choice.framing.locked-scene { text: "FRAMING: locked camera. Hold the selected reference setup as the exact complete frame, preserving its composition, crop, scale, camera height, angle, and lens feel. No handheld drift, push, pan, tilt, zoom, reframe, orbit, or transitional camera movement."; }
+ text-template.street-interview-v1.choice.framing.soft-handheld { text: "FRAMING: realistic tiny handheld sway and operator breathing inside the selected camera setup while preserving its composition and camera side. Never turn that movement into a new crop, angle, camera position, or transition between setups."; }
+ text-template.street-interview-v1.choice.framing.subject-push { text: "FRAMING: make one modest push toward the active speaker inside the selected camera setup, then settle back to that same supplied composition. Never create another angle or camera position."; }
text-template.street-interview-v1.block.pacing { kind: axis; order: 80; parameter: pacing; }
text-template.street-interview-v1.choice.pacing.measured { text: "PACING: measured question and answer delivery with a natural beat between thoughts."; }