Short Video BGM Studio

The Short Video BGM Studio addresses the need for background music in short video creation, where traditional methods of sourcing or commissioning music can be time-consuming and expensive. It generates original instrumental music based on scene descriptions, ensuring a tight match with the video's emotional tone.

The Problem It Solves

In short video production, background music is crucial for setting mood and enhancing viewer engagement. However, finding music that perfectly aligns with specific video content and emotions is challenging. Creators often spend excessive time searching stock libraries or commissioning composers, leading to delays and higher costs. This skill automates music creation by generating original tracks from textual descriptions, eliminating the need for manual selection or custom composition.

How the Skill Works

The core capability is transforming scene descriptions into original instrumental music. The workflow involves:
1. Input Preparation: Users provide a scene description and desired duration. By default, the output is instrumental (instrumental: true), without vocals, and structured to accommodate voiceovers.
2. Music Card Generation: The system locks the scene and duration, then creates a music card specifying use case, mood, genre, rhythm, instruments, emotional arc, and target duration. This ensures the generated music aligns with the video's style.
3. Prompt Construction and Submission: Based on the music card, a positive prompt is constructed, integrating all elements into coherent instructions. Using the bundled MCP client (e.g., scripts/mcp_client.py), the generation task is submitted with a specified model (e.g., suno-5.5).
4. Asynchronous Processing and Delivery: Tasks are submitted asynchronously, and users must poll for status until completion. Delivery includes the audio file with metadata such as duration, MIME type, size, and a review against the music card.

Applicability and Caveats

While the skill suits various scenarios, it has clear boundaries:
- Non-Applicable Cases: For songs with vocals, custom lyrics, covers, or special occasions, other skills like beatra-ai-music-creator or ai-song-cover-studio should be used.
- Importance of Scene Descriptions: The skill relies on detailed scene inputs; vague descriptions may yield mismatched results.
- Payment Confirmation: Before initiating paid tasks, users must confirm the prompt, model options, and costs to avoid unintended charges.
- Fault Recovery: If task IDs are lost, the system attempts recovery via searches or retries, but users should follow standard procedures without fabricating new IDs.