# Speechmatics MCP for AI Agents AI Agent Connect > Speechmatics lets you automate speech-to-text and text-to-speech workflows. Use it to transcribe audio files, generate natural human voices, and manage transcription jobs directly through your AI agent. ## Overview - **Category:** productivity - **Price:** Free - **Endpoint:** https://edge.vinkius.com/vk_preview_FRLuaMwEkSVVnPCae4yQUSbZdPUlWPw9NvqN4KDY/ai-agent-connect - **Tags:** speech-to-text, transcription, text-to-speech, audio-processing, natural-language-processing, voice-synthesis ## Description You can stop wasting time manually typing out meeting notes or trying to figure out how to add subtitles to your videos. This Connector handles the heavy lifting of audio processing, turning spoken words into editable text and vice versa. If you have a long podcast or a series of recorded interviews, you can just point your AI agent at the files and get a transcript back in minutes. It handles different file formats and even lets you choose specific voices for speech synthesis if you need to turn a blog post into an audio file. Because you can find this on the Vinkius catalog, you can swap between different tasks without worrying about the underlying infrastructure. It's built for people who need high-accuracy results without having to build their own audio processing pipeline. You get to focus on the content while the tool handles the data. Whether you are looking to scale your content production or just need a way to make your archives searchable, this connection gives you a direct line to professional grade audio tools. You don't have to worry about API limits or complex configurations; you just describe what you need to your agent and let it do the work. It removes the friction of moving between different audio tools and lets you keep your entire project in one place. You get to spend your time on the creative side of things while the technical heavy lifting happens in the background. ## Tools ### create_temp_key Generate a short-lived key for secure client access. This keeps your main credentials safe while giving your agent the access it needs. ### delete_job Remove a specific transcription job from your active list. Use this to clean up your workspace or cancel a task that was started by mistake. ### generate_tts Turn text into audio using voices like Sarah or Theo. This lets you create high-quality voiceovers for videos or apps without hiring an actor. ### get_job Check the current status and details of a specific job. You can use this to see if your transcription is finished or if it encountered any issues. ### get_transcript Pull the finished text from a completed transcription job. This gives you the final text in formats like SRT or JSON for your projects. ### get_usage See how much of your account quota you have used so far. This helps you track your costs and stay within your monthly limits. ### list_jobs View a list of your most recent transcription tasks. You can quickly see what you've processed recently without searching through logs. ### create_job Start a new batch transcription for an audio file or URL. This lets you process large files in the background so you can keep working on other tasks. ## Prompt Examples **Prompt:** ``` Transcribe the audio file at this URL: https://example.com/audio.mp3 ``` **Response:** ``` I've started a new transcription job for that audio file. I'll use the default English configuration. You can check the status or get the transcript once it's finished. ``` **Prompt:** ``` Generate an audio file of Sarah saying 'Welcome to the future of speech technology'. ``` **Response:** ``` Generating speech with Sarah's voice... Done! You can now access the synthesized audio for the text provided. ``` **Prompt:** ``` List my 5 most recent transcription jobs. ``` **Response:** ``` Fetching your recent jobs... I found 5 jobs. The most recent one is 'Meeting_Notes.mp3' which is currently completed. ``` ## Capabilities ### Transcribe large audio files Convert audio from URLs or base64 data into accurate text for your projects. ### Generate natural human voices Turn written text into high-quality speech using a variety of distinct voice profiles. ### Export transcripts in multiple formats Get your finished text in JSON, plain text, or SRT subtitle files. ### Monitor transcription progress Track the status of multiple active jobs to see when your files are ready. ### Create secure temporary keys Generate short-lived access tokens to keep your main credentials safe. ## Use Cases ### Automated YouTube Subtitles A creator uploads a vlog link and asks the agent to generate a full SRT subtitle file using create_job and get_transcript. ### App Voice Synthesis A developer wants to turn a series of app notifications into audio and uses generate_tts to create natural-sounding voiceovers. ### Archive Searchability A researcher has a folder of old interviews and uses the agent to transcribe them all into searchable text files for analysis. ### Podcast Content Repurposing A podcaster takes a text summary of their episode and asks the agent to generate a narrated news feed using different voices. ## Benefits - Get SRT files for video subtitles automatically using create_job and get_transcript. - Create high-quality voiceovers for scripts using generate_tts with specific voice profiles. - Manage large-scale audio batches without manual monitoring thanks to list_jobs. - Keep your credentials safe by generating short-lived access tokens with create_temp_key. - Save time on data entry by converting meeting recordings into searchable text via get_transcript. - Track your monthly spending and limits easily with get_usage. ## How It Works The bottom line is you get professional audio processing without managing any of the backend complexity. 1. Connect the Speechmatics MCP to your AI client via the Vinkius catalog. 2. Provide your Speechmatics API key in the configuration settings. 3. Ask your agent to transcribe an audio file or generate a voiceover. ## Frequently Asked Questions **Can Speechmatics transcribe long audio files for me?** Yes, it is designed to handle large audio files by creating batch transcription jobs. This allows you to process long podcasts or hours of meetings without worrying about timeouts. **Does Speechmatics support SRT files for video subtitles?** Yes, you can export completed transcripts into SRT format. This makes it easy to add accurate captions to your videos quickly. **What kind of voices can I use for text-to-speech?** It includes several natural-sounding human voices like Sarah, Theo, Megan, and Jack. You can choose the one that best fits your project's tone. **How do I keep my API keys safe when using this Connector?** You can use the tool to generate temporary keys. This gives your AI agent the access it needs for a short period without exposing your main credentials. **Is it good for converting meeting recordings into searchable text?** Yes, it's a great way to turn recorded calls or meetings into plain text or JSON. This makes your archives searchable and easy to summarize. **Can I process multiple audio files at the same time?** Yes, the batch transcription tools allow you to submit multiple files. You can then monitor the progress of all your jobs in one list. **What formats can I get my transcripts in?** You can use the `get_transcript` tool to retrieve results in `json`, `txt`, or `srt` (subtitle) formats. Simply specify the `format` parameter when calling the tool. **Which voices are available for Text-to-Speech?** The `generate_tts` tool supports four high-quality voices: `sarah`, `theo`, `megan`, and `jack`. You can choose the one that best fits your content's tone. **How do I check if my transcription job is finished?** Use the `get_job` tool with your specific `job_id`. It will return the current status (e.g., running, completed) and metadata about the processing task.