# AssemblyAI MCP for AI Agents AI Agent Connect > AssemblyAI MCP lets you turn audio and video files into structured text. It handles high-fidelity transcription, speaker identification, and automatic summaries. You can feed URLs directly to your agent to get back organized meeting notes, podcast transcripts, or sentiment analysis without manual uploading. ## Overview - **Category:** ai-frontier - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_IctBwuEI5fLw0DEGx6JV9ytMfgfL9zH337qQ7vjt/ai-agent-connect - **Tags:** speech-to-text, transcription, audio-intelligence, natural-language-processing, speech-recognition, video-analysis ## Description Stop wasting hours manually scrubbing through audio files or trying to figure out who said what in a messy recording. This Connector gives your AI agent the ability to listen to your media and turn it into actionable data. Instead of you sitting there with a recorder and a notepad, you can just point your agent at a URL. It handles the heavy lifting of turning speech into text while keeping track of different voices. It even figures out the vibe of a conversation, like whether a customer was frustrated or happy, and breaks down long videos into logical chapters. It makes your media library searchable and organized. If you're looking for a way to manage these workflows across multiple apps, Vinkius makes it easy to plug this into your existing setup. You get to move from listening to files to reading insights instantly. It's about getting the gist of a two-hour meeting in ten seconds or turning a raw podcast recording into a polished blog post without the tedious middle steps. ## Tools ### delete_transcript Removes a specific transcription from your history. Use this to keep your workspace clean. ### get_chapters Pulls out the automatic chapters from a video or audio file. It helps you navigate long content quickly. ### get_sentiments Retrieves the emotional tone of a transcript. This is great for gauging customer mood in support calls. ### get_speakers Gets the labels for different voices in a recording. It makes it easy to see who said what in a group meeting. ### get_summary Provides a high-level overview of a transcript. Use this to get the main points of a meeting in seconds. ### get_topics Lists the main subjects discussed in a recording. It helps you find specific parts of a long conversation. ### get_transcript Shows the current status or the final text of a transcription job. Use it to check if your file is ready. ### list_transcripts Shows a list of your recent transcription jobs. This helps you keep track of multiple files at once. ### transcribe_audio_url Sends a link to AssemblyAI to start the transcription process. It's the primary way to get text from a URL. ## Prompt Examples **Prompt:** ``` Transcribe this podcast URL: 'https://example.com/audio.mp3' and enable speaker diarization. ``` **Response:** ``` Transcription started! I've submitted the URL to AssemblyAI with speaker labels enabled. The job ID is 'tr_123'. I'll monitor the processing status and notify you as soon as the high-fidelity transcript is ready. ``` **Prompt:** ``` Show my 5 most recent transcription jobs. ``` **Response:** ``` I've retrieved your recent jobs. You have 5 completed transcripts, including 'Team Weekly Sync' and 'Product Interview'. Would you like to extract the high-fidelity summary for the sync meeting? ``` **Prompt:** ``` Get the sentiment analysis for transcript 'tr_123'. ``` **Response:** ``` Accessing audio intelligence... Transcript tr_123 shows a primarily 'Positive' sentiment with peaks of 'Neutral' during the technical breakdown. I've retrieved the high-fidelity timestamped segments for you. Need more insights? ``` ## Capabilities ### Transcribe audio links Send a URL to your agent and get a full text transcript back in minutes. ### Label different speakers Automatically identify and label different voices in a recording. ### Generate meeting summaries Get a concise overview of a meeting or video without watching the whole thing. ### Extract sentiment data See the emotional tone of a conversation to gauge customer satisfaction. ### Create video chapters Break down long videos into logical segments for easier navigation. ### Identify key topics See what was actually discussed in a recording to find specific information quickly. ## Use Cases ### Podcast Content Creation A creator gives an agent a link and asks for a summary and chapters for a new episode to post on social media. ### Sales Training Analysis A manager asks the agent to find all negative sentiments in last week's sales calls to identify coaching opportunities. ### Meeting Minutes Automation A team lead tells the agent to list the speakers and summarize the final decisions from a sync meeting. ### Video SEO Optimization A YouTuber asks the agent to generate chapters for a 30-minute tutorial video to improve viewer retention. ## Benefits - Stop manual typing by using transcribe_audio_url to turn links into text automatically. - Get the TL;DR of long meetings with get_summary to save time on re-watching hours of footage. - Know exactly who said what with get_speakers for accurate meeting notes and speaker identification. - Gauge customer mood instantly using get_sentiments on support recordings to find pain points. - Organize your content library with get_chapters for easier navigation of long-form video. - Find specific moments in a recording with get_topics for faster searching of key information. ## How It Works The bottom line is you turn raw audio into organized text and data using natural language. 1. Get your API key from the AssemblyAI dashboard. 2. Connect this Connector to your AI client via Vinkius. 3. Provide a URL to your agent to start the transcription. ## Frequently Asked Questions **Can AssemblyAI MCP transcribe YouTube videos?** Yes, if you provide a valid audio URL, the Connector can convert that content into a full transcript for you. **How does it handle multiple people talking?** It uses speaker diarization to identify different voices and label them in the final transcript. **Can it summarize a long meeting?** Yes, you can ask your agent to pull a summary from any completed transcription job to get the highlights. **Is it good for sentiment analysis?** It's excellent for identifying the emotional tone of a conversation, making it useful for analyzing customer feedback. **Can I see the main topics of a recording?** Yes, the Connector can detect and list the primary subjects discussed in your audio files. **How do I get started with AssemblyAI?** You just need an API key from the AssemblyAI dashboard to connect it to your AI client through Vinkius. **How do I find my AssemblyAI API Key?** Log in to your account, navigate to the **API Keys** section in your dashboard, and copy your personal token. **Can I automatically summarize audio via AI?** Yes! Enable the `summarization` option when transcribing, then use the `get_summary` tool to retrieve the high-fidelity AI synopsis. **How do I check transcription status?** Use the `get_transcript` tool with the job ID provided at submission to monitor the status (Queued, Processing, Completed) in real-time.