VideoGen

Voiceover to video

voiceover_to_video

Use when the user already has a recorded voiceover and wants a finished video around that audio. Build an editable video with AI visuals and captions from the uploaded audio file. Fits recorded product pitches, explainers, and training narration. Use script_to_video when the source is written narration rather than audio. Creates a new project and uses team credits. Poll get_workflow_run for progress and downloadUrl after auto-export.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "fileId": "<fileId>"
    },
    "name": "voiceover_to_video"
  }
}

Input Schema

Name	Required	Description
`style`	No	Visual style for generated images: a full, strict paragraph describing medium, texture, palette, and composition. Copy a full app default from get_workflows_guidance, or describe a custom look. Omit for the Realistic default. Image models can add extra objects, text, charts, diagrams, tables, and labels. To keep the picture simple, include this composition lock verbatim: A clear uncluttered subject centered in the frame, occupying only the middle half of the image, with generous empty margins on all four sides, no background clutter. No on-image text, letters, labels, captions, charts, diagrams, tables, legends, or infographic layout unless the user explicitly asked for one specific word or number on screen.
`fileId`	Yes	Uploaded voiceover audio file id.
`quality`	No	Generation quality. Omit to use workspace settings.
`language`	No	Output language as a BCP-47 code, such as en, es, or fr.
`autoExport`	No	When true (the default), the run stays in progress until an MP4 is ready. Use downloadUrl from the result. Set false only if you will call remix_project and then export_project yourself.
`aspectRatio`	No	Output aspect ratio as width:height units (not pixels). Example: { width: 16, height: 9 }.

Output Schema

Name	Required	Description
`error`	No	Error details when status is failed; otherwise null.
`status`	No	Job status: pending, running, succeeded, failed, or cancelled.
`exportId`	No	Export id when autoExport succeeded (vg_expo_...).
`projectId`	Yes	Project created for this exact workflow attempt (vg_proj_...). A retry creates a different project.
`projectUrl`	Yes	Deep link to the project created for this exact workflow attempt. Keep it paired with workflowRunId; a retry returns a new URL.
`downloadUrl`	No	Signed MP4 download URL when autoExport succeeded. Give this to the user.
`attemptIndex`	No	Current or latest attempt index.
`exportFileId`	No	File id of the rendered MP4 when autoExport succeeded.
`workflowType`	No	Workflow type (present after polling).
`workflowRunId`	Yes	Workflow run id (vg_work_...). Keep it paired with this response's projectId and projectUrl; a retry returns a new tuple.
`remixActionIds`	No	Remix action ids from the start response (empty when none requested).
`progressPercentage`	No	Completion progress 0-100 (present after polling).
`downloadUrlExpiresAt`	No	Unix expiry for downloadUrl.

Schema Changelog

Changes observed during successful MCP inspections.

Changed9 schema fields changedNov 4, 2026

removedOutput schema / properties / downloadUrl / anyOf

Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]

addedOutput schema / properties / downloadUrl / type
```
Added value: +[
+  "string",
+  "null"
+]
```

removedOutput schema / properties / downloadUrlExpiresAt / anyOf

Removed value: -[
-  {
-    "type": "number"
-  },
-  {
-    "type": "null"
-  }
-]

addedOutput schema / properties / downloadUrlExpiresAt / type
```
Added value: +[
+  "number",
+  "null"
+]
```

changedOutput schema / properties / error / anyOf

Previous value: -[
-  {
-    "additionalProperties": {},
-    "properties": {
-      "code": {
-        "anyOf": [
-          {
-            "type": "string"
-          },
-          {
-            "type": "null"
-          }
-        ],
-        "description": "Machine-readable error code."
-      },
-      "internalErrorCode": {
-        "anyOf": [
-          {
-            "type": "string"
-          },
-          {
-            "type": "null"
-          }
-        ],
-        "description": "Opaque code for support."
-      },
-      "message": {
-        "description": "Human-readable error description.",
-        "type": "string"
-      },
-      "requirement": {
-        "anyOf": [
-          {
-            "additionalProperties": {},
-            "properties": {
-              "details": {
-                "additionalProperties": {
-                  "type": "string"
-                },
-                "propertyNames": {
-                  "type": "string"
-                },
-                "type": "object"
-              },
-              "type": {
-                "description": "Requirement type (e.g. purchase_add_on).",
-                "type": "string"
-              }
-            },
-            "required": [
-              "type"
-            ],
-            "type": "object"
-          },
-          {
-            "type": "null"
-          }
-        ],
-        "description": "What is needed to resolve the error, when applicable."
-      }
-    },
-    "required": [
-      "message"
-    ],
-    "type": "object"
-  },
-  {
-    "type": "null"
-  }
-]New value: +[
+  {
+    "additionalProperties": {},
+    "properties": {
+      "code": {
+        "description": "Machine-readable error code.",
+        "type": [
+          "string",
+          "null"
+        ]
+      },
+      "internalErrorCode": {
+        "description": "Opaque code for support.",
+        "type": [
+          "string",
+          "null"
+        ]
+      },
+      "message": {
+        "description": "Human-readable error description.",
+        "type": "string"
+      },
+      "requirement": {
+        "anyOf": [
+          {
+            "additionalProperties": {},
+            "properties": {
+              "details": {
+                "additionalProperties": {
+                  "type": "string"
+                },
+                "propertyNames": {
+                  "type": "string"
+                },
+                "type": "object"
+              },
+              "type": {
+                "description": "Requirement type (e.g. purchase_add_on).",
+                "type": "string"
+              }
+            },
+            "required": [
+              "type"
+            ],
+            "type": "object"
+          },
+          {
+            "type": "null"
+          }
+        ],
+        "description": "What is needed to resolve the error, when applicable."
+      }
+    },
+    "required": [
+      "message"
+    ],
+    "type": "object"
+  },
+  {
+    "type": "null"
+  }
+]

removedOutput schema / properties / exportFileId / anyOf

Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]

addedOutput schema / properties / exportFileId / type
```
Added value: +[
+  "string",
+  "null"
+]
```

removedOutput schema / properties / exportId / anyOf

Removed value: -[
-  {
-    "type": "string"
-  },
-  {
-    "type": "null"
-  }
-]

addedOutput schema / properties / exportId / type
```
Added value: +[
+  "string",
+  "null"
+]
```

Changed1 schema field changedOct 16, 2026

changedInput schema / properties / style / description

Previous value: -"Visual style for generated images. Write a full, strict paragraph in the same form as the app defaults below: name the medium, texture, and palette, then lock composition. Do not pass a short label such as \"watercolor\", \"flat art\", or \"cinematic\".\nImage models fill the frame with extra objects, readable text, charts, diagrams, tables, and labels unless the style forbids that. The still then looks crowded and hard to use. Every style must keep the picture simple. Include this composition lock verbatim: A clear uncluttered subject centered in the frame, occupying only the middle half of the image, with generous empty margins on all four sides, no background clutter. No on-image text, letters, labels, captions, charts, diagrams, tables, legends, or infographic layout unless the user explicitly asked for one specific word or number on screen.\nCopy an app default in full (those already include the uncluttered-subject lock), then add the no-text / no-diagram sentence if it is missing. A custom style is allowed only when it is equally long and strict, and includes the same composition lock. Omit this field for the Realistic default.\nWatercolor: Loose watercolor illustration, visible brushstrokes, soft color bleeds, paper texture, muted palette. A clear uncluttered subject centered in the frame, occupying only the middle half of the image, with generous empty margins on all four sides, no background clutter.\nRealistic: Photorealistic photograph, natural lighting\nWhiteboard: Minimalist whiteboard explainer style, simple line drawings, marker sketch aesthetic, clean white background, subtle accent colors. A clear uncluttered subject centered in the frame, occupying only the middle half of the image, with generous empty margins on all four sides, no background clutter.\n3D cartoon: 3D cartoon render, rounded forms, soft lighting, vibrant colors, clean matte materials, smooth stylized characters\nFlat art: Corporate memphis flat art illustration, simple geometric shapes, bold colors, clean composition, white background with sparse subtle geometric accents such as small dots, lines, or shapes scattered in the margins. A clear uncluttered subject centered in the frame, occupying only the middle half of the image, with generous empty margins on all four sides, no busy background patterns.\nPaper: Paper collage illustration, torn edges, layered cut paper shapes, mixed media texture, handmade craft aesthetic, matte paper finish. A clear uncluttered subject centered in the frame, occupying only the middle half of the image, with generous empty margins on all four sides, no background clutter.\nInk paint: Traditional Japanese sumi-e ink wash painting, expressive black ink brushstrokes, varied tonal gradations from deep black to soft gray, rice-paper texture, minimalist composition. No text, letters, calligraphy, kanji, signatures, or red seal stamps unless explicitly required by the subject. A clear uncluttered subject centered in the frame, occupying only the middle half of the image, with generous empty margins on all four sides, no background clutter.\nAnime: Anime illustration, cel-shaded style, vibrant colors, clean linework, soft gradient backgrounds\nEditorial: Editorial illustration, conceptual art, bold geometric shapes, sophisticated color palette, negative space, minimalist composition, strong silhouette\nIsometric: Isometric 3D illustration, clean vector style, soft gradient background, matte pastel colors, simple geometric objects, no characters. A clear uncluttered subject centered in the frame, occupying only the middle half of the image, with generous empty margins on all four sides, no background clutter.\nClaymation: Claymation style, soft clay figurines, plasticine texture, handmade stop-motion aesthetic, warm studio lighting, shallow depth of field\nEducation: Educational infographic style, simple icons, pastel colors, white background. A clear uncluttered subject centered in the frame, occupying only the middle half of the image, with generous empty margins on all four sides, no background clutter.\nChalkboard: Simple, minimalist, white chalk line drawings on dark green chalkboard, hand-drawn sketch style, chalk dust texture. A clear uncluttered subject centered in the frame, occupying only the middle half of the image, with generous empty margins on all four sides, no background clutter.\nBoho: Simple graphic with only a few elements. Boho linocut poster. Textured organic strokes with crisp lines and a white textured background. A clear uncluttered subject centered in the frame, occupying only the middle half of the image, with generous empty margins on all four sides, no background clutter.\nPoster: Risograph print aesthetic, halftone texture, limited duotone palette, paper grain, high contrast, bold flat colors"New value: +"Visual style for generated images: a full, strict paragraph describing medium, texture, palette, and composition. Copy a full app default from get_workflows_guidance, or describe a custom look. Omit for the Realistic default.\nImage models can add extra objects, text, charts, diagrams, tables, and labels. To keep the picture simple, include this composition lock verbatim:\nA clear uncluttered subject centered in the frame, occupying only the middle half of the image, with generous empty margins on all four sides, no background clutter. No on-image text, letters, labels, captions, charts, diagrams, tables, legends, or infographic layout unless the user explicitly asked for one specific word or number on screen."

First observedOct 9, 2026

TDQS

A4.6/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations only declare it is a non-destructive, non-idempotent write with no open-world access. The description adds material side effects the annotations cannot convey: a new project is created, team credits are consumed, and the run is asynchronous with auto-export on by default, requiring get_workflow_run polling for progress and downloadUrl.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Five short sentences ordered as trigger → output → fit cases → alternative → side effects and follow-up, each carrying distinct information. The 'Fits recorded product pitches, explainers, and training narration' sentence is the only borderline filler, but it aids routing.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given six parameters, a nested object, an output schema, and an asynchronous long-running workflow, the description still covers cost, project creation, the export default, and the polling follow-up. Nothing an agent needs to invoke it correctly or handle the result is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100% and the style/fileId/quality/autoExport/aspectRatio descriptions are already detailed, so the description only echoes the audio-file input and the auto-export behavior. Baseline 3 is appropriate since the schema does the heavy lifting and the description adds no syntax or format guidance beyond it.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Build an editable video with AI visuals and captions from the uploaded audio file') and names the input that distinguishes it from neighbors (a recorded voiceover). The agent can tell instantly what comes out and what goes in.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Front-loads the trigger condition ('Use when the user already has a recorded voiceover') and gives an explicit exclusion plus the correct alternative ('Use script_to_video when the source is written narration rather than audio'). It even lists concrete fit cases (product pitches, explainers, training narration).

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All VideoGen tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.