# Resemble AI MCP for AI Agents AI Agent Connect > Resemble AI MCP lets your AI agent generate high-fidelity synthetic speech, clone voices, and manage audio projects directly. It handles everything from text-to-speech with SSML to speech-to-speech transformations and deepfake detection for content security. ## Overview - **Category:** image-video - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_lELKY7eitB2gZRvJGRTlTOUHxzdfjT5MrnBpOAJW/ai-agent-connect - **Tags:** voice-cloning, text-to-speech, synthetic-media, speech-synthesis, audio-processing, deepfake-detection ## Description You can now talk to your AI agent to handle your entire audio production pipeline without jumping between different browser tabs. Instead of manually uploading files and clicking through complex dashboards, you just tell your agent what you need. Whether you're trying to turn a script into a professional voiceover or need to swap a voice while keeping the original emotion and timing, this Connector handles the heavy lifting. It lets you organize your assets into projects, manage your custom voice profiles, and even handle the security side of things like watermarking and deepfake checks. It's a massive time saver if you're tired of the back-and-forth between your creative tools and your project management. You'll find this sitting in the Vinkius catalog alongside hundreds of other specialized connectors, making it easy to plug into your existing workflow. It turns a multi-step production process into a simple conversation with your agent. This means you can focus on the creative side of your work while the agent handles the technical tasks of voice synthesis, project organization, and content verification. You get a direct line to high-fidelity audio tools that usually require a lot of manual clicking and file management. ## Tools ### add_watermark Applies an imperceptible watermark to an audio file. This helps you protect your assets and verify ownership. ### create_clip Generates a new text-to-speech clip using SSML. You can control the exact timing and emotion of the output. ### create_project Sets up a new project folder for your audio assets. This keeps your different campaigns organized. ### create_recording Uploads an audio recording to train a new custom voice. This is the first step in voice cloning. ### create_voice Creates a new custom voice profile in your account. Use this to build a unique brand identity. ### delete_voice Removes a custom voice from your library. Use this to clean up old profiles or test results. ### detect_deepfake Checks if an audio clip is real or AI-generated. This is essential for verifying content authenticity. ### get_clip Retrieves details for a specific audio clip. Use this to check the status or metadata of a file. ### get_voice Fetches the details of a specific voice profile. This helps you see the settings for your custom voices. ### list_clips Shows all clips contained within a specific project. This helps you see everything in a campaign at once. ### list_projects Shows all projects currently in your account. Use this to navigate between different creative works. ### list_recordings Lists the recordings used to train a specific voice. This helps you track your training data. ### list_voices Shows all available custom and system voices. Use this to see your full library of options. ### speech_to_speech Transforms an input audio file into a target voice. This keeps the original emotion while changing the speaker. ### update_clip Modifies the details of an existing audio clip. Use this to change settings on a clip you already made. ### verify_watermark Checks an audio file to see if it has a watermark. This confirms your content is protected. ## Prompt Examples **Prompt:** ``` I need to make a voiceover for a new marketing script. Can you create a clip in my 'Summer Campaign' project using the 'Narrator' voice? ``` **Response:** ``` Generating your voiceover now... **Success!** I've created the clip in your **Summer Campaign** project using the **Narrator** voice. | Project | Voice | Status | | :--- | :--- | :--- | | Summer Campaign | Narrator | ✅ Completed | Your new clip is ready for review. Would you like me to list all clips in this project for you? ``` **Prompt:** ``` Can you check if this audio file is a deepfake? Here is the link: https://example.com/audio.mp3 ``` **Response:** ``` Running deepfake detection analysis... **Analysis Complete:** The audio file provided has a **98% probability** of being synthetic (AI-generated). **Summary:** - **Source:** https://example.com/audio.mp3 - **Confidence:** High - **Verdict:** Likely Synthetic Would you like me to check any other files? ``` **Prompt:** ``` Show me all the voices I have available and tell me which ones are custom. ``` **Response:** ``` Here are the voices currently available in your account: **Custom Voices:** - *Narrator* (Custom) - *Promo_Voice_01* (Custom) - *Storyteller_V2* (Custom) **System Voices:** - *Default_Male_01* - *Default_Female_01* - *UK_Accent_Standard* Which voice would you like to use for your next project? ``` ## Capabilities ### Generate high-fidelity audio clips from text Create audio from scripts using SSML for precise timing and emotion. ### Convert source audio to a target voice Swap a voice in an existing recording while preserving the original performance. ### Organize audio assets into projects Group clips and files into specific projects for better management. ### Create and manage custom voice profiles Build unique voices and manage the recordings used to train them. ### Detect deepfakes and verify authenticity Check if an audio file is real or AI-generated to ensure content security. ### Apply imperceptible watermarks Add digital watermarks to audio files to track and protect your content. ## Use Cases ### Localizing content for global audiences A YouTuber needs to localize a video into five languages. They ask the agent to use create_clip for each script using a specific voice. ### Swapping voices in game dialogue A game studio wants to change a character's voice but keep the performance. They use speech_to_speech to swap the voice while keeping the original acting. ### Verifying leaked audio for deepfakes A security firm needs to check if a leaked audio clip is a deepfake. They provide the URL to the agent to run detect_deepfake. ### Organizing large-scale marketing campaigns A marketing team needs to organize a huge campaign of voiceovers. They use create_project and list_projects to group all assets by campaign name. ## Benefits - Create high-quality audio clips using create_clip without leaving your chat, allowing for much faster content iteration. - Save hours on audio production by using speech_to_speech to swap voices while keeping the original emotion and timing. - Keep your assets organized by using create_project and list_clips to manage large libraries of audio files. - Build a unique brand identity by creating custom voices with create_voice and training them with create_recording. - Protect your content from misuse by using add_watermark and verify_watermark to secure your synthetic audio. - Ensure content authenticity by using detect_deepfake to identify synthetic manipulations in audio files. ## How It Works The bottom line is you get a conversational interface for a professional-grade voice synthesis studio. 1. Connect your Resemble AI API token to the Connector. 2. Tell your AI agent to create a voice, a project, or a clip. 3. Receive the audio file or confirmation directly in your chat. ## Frequently Asked Questions **Can the Resemble AI MCP create realistic voices?** Yes, it uses high-fidelity synthesis to create voices that sound natural. You can even train custom voices using your own recordings to get a specific sound. **How does the Resemble AI MCP handle deepfakes?** It includes a detection tool that analyzes audio files to determine if they were generated by AI. This helps you verify the authenticity of content quickly. **Can I use the Resemble AI MCP to clone my own voice?** You can. By using the recording tools, you can upload your own audio to train a custom voice profile that sounds like you. **Does the Resemble AI MCP support SSML tags?** Yes, the clip generation tool supports SSML, which allows you to have precise control over the timing, pitch, and emotion of the synthetic speech. **Can I organize my audio files with the Resemble AI MCP?** Absolutely. You can create specific projects to group your clips, making it much easier to manage large marketing campaigns or game dialogue. **How do I protect my audio from being stolen using the Resemble AI MCP?** You can use the watermark tool to apply an imperceptible digital mark to your audio files. This helps you verify ownership and protect your content. **Can I use SSML tags when creating an audio clip?** Yes! The `create_clip` tool fully supports SSML in the body field, allowing you to control pitch, rate, and emphasis for more natural-sounding speech. **How can I transform an existing audio file into a different voice?** Use the `speech_to_speech` tool. Provide the target `voice_uuid` and a Base64 encoded audio sample or URL, and the AI will re-voice the content while keeping the original performance characteristics. **Is there a way to check if an audio file is a deepfake?** Absolutely. The `detect_deepfake` tool allows you to submit an audio URL to Resemble's security engine to analyze whether the speech was generated by AI.