Text to speech API and MCP: voices for your AI agent

    Turn scripts into voice-overs, dialogue and sound effects from your AI agent with ElevenLabs, Fish Audio, MiniMax, Google Gemini, HeyGen and StepFun through one MCP connection. You need no API keys or voice platform accounts. Your agent runs the tool from your balance and tells you what it cost. Transcription is here too.

    22 tools and 149 skills that use them. Pay per use from one balance, no API keys.

    Connect your AI AgentSee the tools

    Starting prices. Prices can vary with length and quality. Your AI Agent works out the exact price, runs the tool from your balance and tells you what it cost. Failed uses are free.

    What do you want your agent to do?

    Our text to speech tools and MCPs

    SourceBest forExample, today's priceOfficial MCP
    ElevenLabsExpressive voice-overs, dialogue with several voices, sound effects, voice changing and transcriptionElevenLabs Text to Speech, $0.052 per 1,000 charactersYes
    Fish AudioA public library of over 2 million voices, word timings for captions, transcripts of up to 60 minutesFish Audio: S2.1 Pro text to speech, $0.0195 per 1,000 charactersYes
    MiniMaxLong scripts, up to 10,000 characters per run, in HD or TurboMiniMax: Speech 2.8 HD, $0.13 per 1,000 charactersYes
    Google Gemini TTS30 prebuilt voices and two-speaker dialogueGoogle: Gemini 3.8 Flash TTS, $0.065 per voice clip
    HeyGenHeyGen voices with word timings for captions, up to 5,000 characters per runHeyGen Text to Speech, $0.0052 per 1,000 charactersYes
    StepFunNarration in six languages with laughter, breaths and natural pausesStepFun Text to Speech, $0.0468 per 1,000 characters

    ElevenLabs, Fish Audio, HeyGen and StepFun list their voices for free, so ask your agent to show you the options before it reads your script.

    How to choose the best text to speech tool or MCP for you

    Expressive or plain

    For ads, stories and characters, ElevenLabs v4 takes audio tags like [whispering], and StepFun adds laughter, breaths and pauses. For plain narration any voice here works, so compare the prices on this page.

    Captions and timing

    Fish Audio's Speech with Word Timings and HeyGen return the timing of each word, so captions line up. ElevenLabs Forced Alignment adds word timings to audio you already have.

    Voices and languages

    Fish Audio's public library holds over 2 million voices, Gemini TTS has 30 prebuilt voices, StepFun speaks six languages including Chinese, Japanese and Korean, and HeyGen detects the language from your text.

    Length per run

    MiniMax takes up to 10,000 characters per run and HeyGen up to 5,000. For an audiobook, your agent splits the script into parts, and ElevenLabs Text to Dialogue can carry the context from one part to the next.

    If you needChooseToday's price
    An expressive voice-over for an adElevenLabs Text to Speech$0.052 per 1,000 characters
    Two or more people talkingElevenLabs Text to Dialogue$0.052 per 1,000 characters
    Speech with word timings for subtitlesFish Audio Speech with Word Timings$0.0195 per 1,000 characters
    A long script in one runMiniMax Speech 2.8 HD$0.13 per 1,000 characters
    A sound effect for a videoElevenLabs Sound Effects$0.156 per minute
    A transcript of a call, meeting or videoElevenLabs Speech to Text$0.0048 per minute
    Clean speech from a noisy recordingElevenLabs Voice Isolator$0.156 per minute

    Quick picks: ElevenLabs for expressive voice-overs and dialogue, Fish Audio for captions and a large voice library, MiniMax for long scripts, and ElevenLabs Speech to Text for transcripts.

    Why use Agensi instead of separate text to speech MCPs?

    ElevenLabs, Fish Audio, MiniMax and HeyGen each run an official MCP server, and each one works with your account, plan or API key at that brand. Using three voice brands the direct way means three accounts and three balances. With Agensi you connect once: every voice here, plus the video and image tools you put the voice on, runs from one balance, and your agent tells you what each use cost.

    What you getOfficial MCP per brandOwn API key or community serverAgensi
    AccountsOne per brandOne per brandOne
    API keysPer brand, or a sign-inPer brand, kept by youNone
    PaymentA plan or credits per brandPer brandOne balance, per use
    Spending controlVariesUp to youLimits per use, day and month, or ask first
    Several sources in one answerConnect several serversSeveral servers and keysOne connection
    Skills for the whole jobNoNoYes, from the marketplace

    When another way is the better choice. If you make most of your audio with ElevenLabs and want its voice design, its agents platform or its newest models first, its own MCP and plan are the better fit. Developers who build speech into an app should call the provider's API directly. Agensi fits when voice is one step in a bigger job, next to video, images or data.

    Set up in your AI agent

    Connect your AI agent to Agensi once, sign in and click Allow. After that it can use every tool: it runs one from your balance and tells you what it cost. No API keys or provider accounts.

    1. Open Settings, then Connectors, then Browse connectors and add Agensi (or Add custom connector and paste https://mcp.agensi.io/mcp/auth).
    2. Sign in to Agensi and click Allow. One agent per account: connecting a new one disconnects the old one.
    3. Ask for what you want, for example: "Use Agensi to turn my 2,000-character podcast intro into a conversation between two ElevenLabs voices, then line the script up with the audio so I get word timings for subtitles." Top up from $5 if your balance is low; failed uses are free.

    Grok, VS Code, GitHub Copilot, Cline, Kiro and other agents with remote MCP and sign-in work too: see the setup guide.

    What it costs: a worked example

    "Use Agensi to turn my 2,000-character podcast intro into a conversation between two ElevenLabs voices, then line the script up with the audio so I get word timings for subtitles."
    About$0.1184

    At today's starting prices. Your agent works out the exact price for your request and tells you what it cost.

    Combine text to speech tools with skills

    Text to speech MCP: frequently asked questions

    Yes. Your agent runs ElevenLabs Text to Speech, Text to Dialogue, Sound Effects and Speech to Text through Agensi, paid from your Agensi balance. You don't need an ElevenLabs plan or API key, and the same balance pays for Fish Audio, MiniMax and the other voices on this page.

    Ask it to list voices first. ElevenLabs, Fish Audio, HeyGen and StepFun list theirs for free, and Fish Audio's public library holds over 2 million. Name the voice you want, or describe it (calm, warm, British) and let your agent search for a match before it reads your script.

    Voice tools are priced per 1,000 characters, per minute of audio or per run, and each tool shows its starting price on this page. Your agent works out the exact price from your script, runs it from your balance and tells you what it cost. Failed uses are free.

    Your agent gives you a link to each file, and the link works for 7 days. The files the tools make also stay in Dashboard → Balance for 30 days, so download the ones you want to keep, or ask your agent to save them where you work.

    Read next

    Browse all tools

    Get in touch

    Questions, partnership ideas, or feedback on the platform? Tell us what's working and what isn't, we love to hear from you!

    SR

    Samuel Rose

    Founder & CEO of Agensi

    Email UsBook a Call

    info@agensi.io

    We typically respond within a few hours during business hours.