# ElevenLabs MCP for AI Agents AI Agent Connect > ElevenLabs MCP connects your audio workspace to your AI client, letting you generate high-quality speech, clone voices, and create sound effects via chat. It handles everything from text-to-speech and voice conversion to managing your entire library of custom voices and dubbing projects. ## Overview - **Category:** ai-frontier - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_bLutSHvKND4deezkU6LVk06zInlnHKVJDQrwFVAI/ai-agent-connect - **Tags:** text-to-speech, voice-cloning, generative-audio, speech-synthesis, ai-voice, audio-processing ## Description ElevenLabs connects your audio workspace to your AI agent so you can handle high-fidelity voice generation without switching tabs. Instead of manually clicking through a dashboard to find the right voice or adjust stability settings, you can just tell your agent to generate a voiceover for a script or find a similar voice to a favorite one. It handles the heavy lifting of audio processing, like isolating a voice from a noisy background or converting one person's speech into another's while keeping the original emotion. You can also manage your entire library of custom voices, create new ones from prompts, and oversee long-form dubbing projects. Using this Connector via the Vinkius catalog means you're moving audio assets into your primary workspace, making it way easier to iterate on content, test different voice personalities, and prototype sound effects for games or videos in real-time. You get a direct line to professional-grade audio tools without leaving your current conversation. ## Tools ### add_dictionary_from_rules Add pronunciation rules to your library to handle specific naming or terms. This improves the accuracy of the speech. ### isolate_audio Strip background noise out of a recording to get a clean vocal track. It makes your audio sound professional. ### convert_speech Change one person's voice into another while keeping the original delivery. This is perfect for voice acting. ### create_dub Start a new dubbing project for translating and localizing video content. It handles the initial setup for your dub. ### create_project Launch a new Studio project to organize your audio work. This keeps your workspace tidy. ### create_single_use_token Generate a one-time token for secure, limited access to specific actions. Use this for extra security. ### delete_history_item Remove a specific entry from your generated audio history. This keeps your history clean. ### delete_voice Permanently remove a custom voice from your library. Use this to clear out old voice clones. ### design_voice Create a brand new voice using a descriptive text prompt. This lets you build unique voices from scratch. ### edit_voice Modify the settings and characteristics of an existing custom voice. Use this to fine-tune your audio. ### find_similar_voices Locate voices in your library that match the style of a specific one. This helps you find the right tone quickly. ### generate_sound Create unique sound effects from a written description. This is great for rapid prototyping of game audio. ### get_character_stats Check how much a specific character or voice has been used. This helps you track usage across different projects. ### get_dub_status Check the progress of your current dubbing projects. This helps you track long-running audio tasks. ### get_history_audio Download the audio file for a specific item in your history. This lets you grab files without manual searching. ### get_history_item Pull the details for a previous audio generation. Use this to see the settings used for a past file. ### get_snapshot Grab a current snapshot of a project's status. This is useful for saving progress on complex audio work. ### get_user Pull your account information and current subscription status. This helps you check your limits. ### get_voice Fetch the specific details of a single voice in your library. Use this to see the exact settings of a voice. ### list_agent_branches See the different branches for a Conversational AI agent. This helps manage complex agent logic. ### list_agents View all the conversational agents you've created. This gives you an overview of your interactive voices. ### add_dictionary_from_file Import a pronunciation dictionary from a PLS file. This ensures the AI pronounces specific terms correctly. ### list_dictionaries Browse all the pronunciation dictionaries in your account. This ensures you know which rules are active. ### list_dubs See all your active and completed dubbing projects. This makes it easy to track your localization work. ### list_history Review the history of all your previous audio generations. This helps you find and reuse old work. ### list_models See the different audio models available for your use. This helps you choose the best model for your task. ### list_voices View every voice currently available in your ElevenLabs library. You can quickly see what you have to work with. ### stream_isolate_audio Get real-time audio isolation for live or streaming content. This works for live audio processing. ### stream_convert_speech Perform live voice conversion for real-time speech-to-speech needs. It keeps the emotion while changing the voice. ### stream_speech Stream text-to-speech audio for low-latency applications. Use this for real-time responses. ### create_designed_voice Save a new voice you've created using specific design parameters. It keeps your custom voices organized. ### create_agent Build a Conversational AI agent for interactive audio experiences. Use this to create custom interactive voices. ### list_projects See a list of all your active Studio projects in one place. This helps you stay organized. ### create_speech Turn a block of text into a high-quality audio file. This is the core tool for generating voiceovers. ## Prompt Examples **Prompt:** ``` I need a voiceover for a 30-second ad about a new coffee brand. Can you generate it using a warm, friendly voice? ``` **Response:** ``` I've generated a warm and friendly voiceover for your coffee brand advertisement. **Audio Details:** - **Voice Style:** Warm & Friendly - **Duration:** ~30 seconds - **Status:** Ready for download Would you like me to adjust the stability or try a different voice style? ``` **Prompt:** ``` Can you find a voice in my library that sounds similar to the 'Rachel' voice but a bit deeper? ``` **Response:** ``` I've searched your library for voices similar to 'Rachel' with a deeper tone. Here are the best matches: - **Clyde** (ID: 2EiwWnXFnvU5JabPnv8n) - **Marcus** (ID: 7m00Tcm4TlvDq8ikWAM) Would you like to hear a sample of one of these? ``` **Prompt:** ``` I have a recording with a lot of wind noise. Can you clean it up for me? ``` **Response:** ``` I've processed your recording using the audio isolation tool. **Changes made:** - **Background Noise:** Removed - **Vocal Clarity:** Enhanced - **Output:** Clean vocal track ready Your audio is much clearer now. Let me know if you'd like any other adjustments. ``` ## Capabilities ### Generate lifelike speech Turn text into high-quality audio with specific control over voice stability. ### Clone and manage voices Create new voices from prompts and manage your existing library through chat. ### Convert speech to speech Change one voice into another while preserving the original emotion and delivery. ### Isolate audio tracks Remove background noise from recordings to get clean vocal tracks. ### Create sound effects Generate unique sound effects from text descriptions for any project. ### Manage dubbing projects Oversee multi-language dubbing workflows and check project status. ## Use Cases ### Rapid Social Media Voiceovers A content creator needs to turn a script into a TikTok voiceover. They ask their agent to use create_speech with a specific voice ID to get a high-quality audio file in seconds. ### Game Audio Prototyping A sound designer needs a 'futuristic door opening' sound. They ask their agent to use generate_sound to create several variations they can test immediately. ### Cleaning Up Field Recordings A podcaster has a recording with heavy wind noise. They ask their agent to use isolate_audio to clean up the vocal track so it's ready for the final edit. ### Multi-Language Dubbing A studio is localizing a short film. They use create_dub to initiate the project and get_dub_status to monitor the progress of several different language tracks at once. ## Benefits - Skip the dashboard by using create_speech to generate voiceovers directly in your chat, saving you from constant tab switching. - Refine your audio quality instantly with isolate_audio to strip out background noise without needing a separate editing suite. - Create unique assets on the fly with generate_sound to prototype game effects or video transitions using simple text prompts. - Maintain consistent branding by using design_voice and edit_voice to manage a library of custom voices that sound exactly right. - Speed up your localization workflow with create_dub and get_dub_status to track your multi-language audio projects in real-time. - Ensure perfect pronunciation for technical terms by using add_dictionary_from_rules to feed specific rules into your audio model. ## How It Works The bottom line is you get a direct command line for every ElevenLabs audio feature inside your favorite AI chat. 1. Subscribe to the ElevenLabs MCP on Vinkius. 2. Paste your ElevenLabs API Key into your AI client settings. 3. Ask your agent to generate audio, manage voices, or start a dubbing project. ## Frequently Asked Questions **How can I use the ElevenLabs MCP to create voiceovers faster?** You can generate high-quality voiceovers directly through your AI chat. Just provide your script and tell the agent which voice to use, and it will handle the generation and delivery for you. **Can the ElevenLabs MCP help with multi-language dubbing?** Yes, it allows you to start and manage dubbing projects for various languages. You can check the status of your dubbing tasks and oversee the entire workflow from your agent. **Does the ElevenLabs MCP support sound effect generation?** It does. You can describe a sound effect in plain text, and the Connector will generate the audio for you, which is great for game development and video production. **Is it possible to clean up noisy audio using the ElevenLabs MCP?** Yes, the Connector includes a tool to isolate audio, which removes background noise from your recordings to provide a cleaner vocal track. **How do I manage my custom voices with the ElevenLabs MCP?** You can list, design, edit, and delete your custom voices through your AI agent. It gives you full control over your voice library without needing to visit the website. **Can I use the ElevenLabs MCP for real-time speech conversion?** Yes, the Connector supports streaming voice conversion, which allows you to change one voice into another in real-time while maintaining the original emotion. **How can I convert text to a specific voice using this server?** You can use the `create_speech` tool. Simply provide the `voice_id` and the `text` you want to synthesize. The agent will generate the audio for you. **Can I see all the voices available in my account?** Yes! Use the `list_voices` query. It will return a list of all available voices, including their IDs, names, and categories, so you can choose the right one for your project. **Is it possible to remove background noise from an existing audio file?** Absolutely. Use the `isolate_audio` tool by providing the audio in base64 format. The server will process it and return a clean version with the background noise removed.