videoagent-audio-studio_skill

Overview

This skill provides one-command access to TTS, music generation, sound effects, and voice cloning so you can produce any audio without juggling multiple API keys. It routes user requests to the best available model (ElevenLabs, fal.ai) and returns ready-to-use audio URLs. Use it to generate narrations, background music, SFX, or clone a voice from a sample quickly and reliably.

How this skill works

The skill inspects the user's intent and selects the optimal backend: ElevenLabs for high-quality TTS, ElevenLabs SFX and voice cloning, and fal.ai (cassetteai-music) for background music and fast SFX. It calls a local or hosted proxy MCP server which forwards requests to the appropriate model, waits for generation, and returns an audio URL. Configuration requires an ElevenLabs API key and optional fal.ai key for music.

When to use it

• You need a narration or voice-over for video, podcast, or tutorial.
• You want background music composed to a prompt and duration.
• You need a short sound effect (door creak, whoosh, ambience).
• You want to clone a voice from an audio sample for consistent narration.
• You need low-latency TTS for near-real-time interactions.

Best practices

• Provide concise prompts with style, mood, and duration for music (e.g., ‘60s lounge, mellow, 90 seconds’).
• Specify voice_id and model_id when you need a particular tone or latency (use turbo model for low latency).
• Keep SFX descriptions focused and set duration between 1–22 seconds for reliable results.
• Supply a clear, noise-free audio sample for voice cloning to improve fidelity.
• Host your own proxy if you need regional reliability or to control API keys.

Example use cases

• Generate a 90‑second ambient music bed for a meditation app via cassetteai-music.
• Produce a professional product launch voice-over using elevenlabs-tts-v3 and return a sharable audio URL.
• Create a 3‑second sci‑fi door opening SFX with elevenlabs-sfx for a game asset.
• Clone a presenter’s voice from a short clip and use it to narrate a multi-episode course.
• Deliver sub-second TTS responses in interactive demos using the turbo low-latency model.