What causes "The transcription asset must be audio or video." in remotion-dev/remotion?

Pass assetPath pointing at an audio or video file (e.g. .mp3, .wav, .mp4, .mov).

The transcription asset must be audio or video.

Error message

The transcription asset must be audio or video.

What it means

The WebMcp transcription tool resolves the asset then checks its preview file type. Whisper can only transcribe audio/video, so assets of other types (images, fonts, JSON, etc.) are rejected with this error.

Source

Thrown at packages/studio/src/components/WebMcp.tsx:506

							},
							noRepeatNgramSize: {type: 'integer', minimum: 0, default: 0},
						},
						additionalProperties: false,
					},
					annotations: {readOnlyHint: false},
					execute: async (input) => {
						if (!isOptionalPackageInstalled(WHISPER_WEBGPU_PACKAGE)) {
							return missingOptionalPackageResult(WHISPER_WEBGPU_PACKAGE);
						}

						const assetPath = resolveAssetPath({
							assetPath: input.assetPath,
							currentContent: currentContentRef.current,
							staticFiles: staticFilesRef.current,
						});
						const fileType = getPreviewFileType(assetPath);
						if (fileType !== 'audio' && fileType !== 'video') {
							throw new Error(
								'The transcription asset must be audio or video.',
							);
						}

						const whisper = await import('@remotion/whisper-webgpu');
						const modelName = input.model ?? 'small.en';
						if (typeof modelName !== 'string') {
							throw new Error('model must be a string.');
						}

						const model = whisper
							.getAvailableModels()
							.find((candidate) => candidate.name === modelName);
						if (!model) {
							throw new Error(`Unknown Whisper model: ${modelName}.`);
						}

						const task = input.task ?? 'transcribe';

View on GitHub (pinned to b2f4e34732)

Solutions

Pass assetPath pointing at an audio or video file (e.g. .mp3, .wav, .mp4, .mov).
Select the correct media asset in Studio before invoking the tool.
If the source is a different format, convert it to audio first.
Verify the file's actual media type matches its extension.

Example fix

// before
await tool.execute({assetPath: 'poster.png'});
// after
await tool.execute({assetPath: 'voiceover.mp3'});

Defensive patterns

Strategy: validation

Validate before calling

const fileType = getPreviewFileType(assetPath);
if (fileType !== 'audio' && fileType !== 'video') {
  throw new Error('Transcription requires an audio or video asset.');
}

Try / catch

try {
  await transcribeTool({assetPath});
} catch (e) {
  if (e.message.includes('must be audio or video')) {
    chooseMediaAssetInstead();
  }
}

Prevention

Check getPreviewFileType before calling.
Keep media assets organized separately in public/.
Never point transcription at images/fonts/data files.

When it happens

Trigger: Calling the WebMcp transcribe tool with assetPath (or a selected Studio asset) pointing to a non-media file in public/ — e.g. a .png, .json, or .ttf file.

Common situations: Agents picking the wrong asset from public/; users selecting a thumbnail or data file in Studio; assets with misleading extensions (e.g. a .txt containing a transcript).

Understand the failure class

Background: "Must be a positive integer", "Invalid value", "Unsupported": the invalid-argument-value error family, when a library rejects the value you pass — this error's family across 35 libraries.

Related errors

AI-assisted analysis of remotion-dev/remotion@b2f4e34732 (2026-09-09). Data as JSON: /api/errors/6a90e66c209fe056. Report an issue: GitHub.