ElevenLabs for AI agents: pricing, alternatives and setup

What is ElevenLabs?

ElevenLabs makes speech models and the products built on them: an AI voice generator, dubbing, an audiobook reader, and a platform for voice agents. Developers reach the same models through its API.

Its text to speech is known for sounding human, with emotion and natural pacing, in many languages. As of November 2026 the lineup is Eleven v4, its most expressive model (released September 2026), v4 Turbo for low latency, and Flash v2.5, a fast conversational model with about 100 ms of model latency in 32 languages. Scribe v2 and Scribe v2 Realtime turn speech back into text. You can clone a voice from a recording, design a new one, or pick from a large shared voice library.

ElevenLabs pricing in 2026

ElevenLabs sells monthly plans for its apps, and bills its API in dollars per character. The plans differ in included credits, commercial rights, voice cloning and how many requests can run at once. Annual billing costs ten months' price for twelve.

From ElevenLabs' pricing page, checked November 2026. In the apps, text to speech uses about 1 credit per character, so about 900 credits per minute of speech at the pace these tables assume.

ElevenLabs API pricing per character and per hour

The API is what Ork uses, and it is billed differently from the apps. Each model has a dollar price per 1,000 characters that is the same on every plan; a subscription includes some usage, and Pay As You Go lets you top up a dollar balance (from $5, valid for 12 months) with or without a plan. Without auto top-up, requests stop at $0 instead of running up a bill. Older guides quoting $0.10 per 1,000 characters predate a 2026 price cut.

From ElevenLabs' API pricing page, checked November 2026. An hour of speech is taken as about 54,000 characters.

What a talking Ork bot costs on ElevenLabs

Ork calls ElevenLabs with your key, so ElevenLabs bills you at its API rates and Ork adds nothing on top. The cost depends on how much your bots say and how much of your speech is transcribed on the server.

The worked example is one bot that reads about 30 minutes of replies aloud each day and transcribes about 30 minutes of what you say, every day for a month: 15 hours of speech (about 810,000 characters) and 15 hours of transcription. On ElevenLabs that comes to $32.40 of Flash v2.5 speech and $3.30 of Scribe v2 transcription: about $36 a month. A light user who hears and speaks 10 minutes a day pays around $12; a bot that talks for two hours a day costs about $130 in speech alone.

Two things lower the bill. In Chrome on a desktop, the mic uses the browser's own speech recognition, so that transcription costs nothing. And a self-hosted Ork can set VOICE_SUMMARY_MIN_CHARS so long replies are read as a short summary instead of in full, which cuts characters at the cost of one extra model call.

Usage prices from each provider's pricing page, checked November 2026. Where speech is bought as a monthly plan, as with Cartesia, the cost shown is the fee for the smallest plan that covers it. Ork doesn't charge for voice: you connect your own key and the provider bills you.

ElevenLabs vs OpenAI, Cartesia, Fish Audio and Deepgram

ElevenLabs is not the cheapest voice. Per character, Flash v2.5 is about the same as Cartesia on its Startup plan and more expensive than Deepgram's Aura-2, while Fish Audio costs about 38% of its price and OpenAI's tts-1 about 38% of it. What you pay for is voice quality, the size of the voice library and cloning that people consistently rate among the best.

Prices in US dollars from each provider's pricing page, checked November 2026. An hour of speech is taken as about 54,000 characters (150 words a minute); real voices run a little faster or slower. Fish Audio bills per UTF-8 byte, which is one byte per character in English.

ElevenLabs alternatives

The best alternative depends on why you're leaving. If it's price, Fish Audio costs about 38% of the per-character price, with cloning and a big voice library, and a self-hosted Kokoro server costs only hardware. If it's latency on calls, Cartesia is built for it, though in Ork it only speaks. If you already pay OpenAI, OpenAI is the quickest start, but its speech model shuts down on January 6, 2027.

Outside Ork, Deepgram is strong at transcription (Nova-3 from $0.26 an hour) with Aura-2 voices that are cheaper than Flash v2.5 but plainer; Google, Azure and Amazon Polly are cheap at scale and come with cloud compliance paperwork; Hume and Inworld focus on expressive or low-cost voices for agents. PlayHT, once a common alternative, shut down at the end of 2025. Ork has no built-in connection for any of these.

What you can do with ElevenLabs in Ork

ElevenLabs both speaks and transcribes, so it covers every voice feature in Ork. The voice and calls guide shows where each control lives.

• Talk instead of typing. Hold the mic in the composer, speak, and let go to send. When Ork transcribes on the server, as it does in a phone browser or the desktop app, it uses ElevenLabs' scribe_v2.
• Hear replies. Choose Speak this reply on any message, or turn on Read replies aloud in a bot's settings to hear everything it says.
• Call a bot. Choose Call at the top of a chat to talk in real time. You can cut in while the bot is speaking, and the transcript stays in the chat.
• Get called. Ask "call me at 5 to go over the day" and the bot rings you then. A bot may also call if you've missed the same reminder several times.
• Give each bot its own voice. The Voice next to Read replies aloud overrides your default for that bot, so your inbox bot and your sales bot don't sound alike.
• Use your own voices. The voice picker lists the voices in your ElevenLabs account, including ones you cloned or saved from its library.
• Send voice notes from chat apps. A voice note sent to a bot through Telegram, for example, is transcribed with ElevenLabs before the bot reads it. See messaging apps.

When to choose ElevenLabs

Choose ElevenLabs when the voice itself matters: you'll listen to a bot every day, you want it to sound like a person, or you want it to sound like a specific person you have permission to clone. It is also the simplest choice for full voice support, since one key speaks and transcribes, and its Scribe transcription is among the cheapest hosted options.

Skip it if cost per hour is what you're optimizing, if you need audio to stay on your own hardware, or if you need data residency or zero retention without an Enterprise contract. The comparison puts every option side by side.

Limits and gotchas

• Ork always speaks with Eleven Flash v2.5 (eleven_flash_v2_5), ElevenLabs' fast model for conversation. You can't switch to a more expressive model such as Eleven v4 from Ork.
• Ork doesn't clone voices itself. Create the clone in ElevenLabs first, then pick it in Ork.
• Each sentence is generated as a finished clip and played in order, rather than streamed word by word, so there is a short pause before the bot starts talking.
• Speech and transcription go through ElevenLabs' servers and use your ElevenLabs account's balance. If it runs out, Ork tells you so instead of failing quietly.
• The Free plan is for non-commercial use with attribution. A bot that talks to customers needs a paid plan.
• Unused plan credits roll over for up to two months, but cancelling or downgrading forfeits them at the end of the cycle. Recent reviews complain about exactly this, so downgrade after you've used them.
• Zero retention mode and data residency in the EU, India or Singapore are Enterprise-only.
• Calls and hold-to-talk are a server setting. On a self-hosted server they stay off until the admin turns them on.
• In a desktop browser such as Chrome, Ork may use the browser's own speech recognition for the mic instead of ElevenLabs.

How to set it up

Anyone can do this in their own settings, on Ork Cloud or a self-hosted server. The key belongs to your account, not the whole server.

1. Create an API key in your ElevenLabs account. Give it permission for text to speech, speech to text and reading voices.
2. In Ork, open Settings → Voice, choose ElevenLabs as the Provider, paste the API key and choose Connect.
3. Pick a Voice and use Hear a sample to check it.
4. Optionally, open a bot's settings, expand Advanced, turn on Read replies aloud and give it its own Voice.
5. If you run your own server and want calls and hold-to-talk, set PHONE_CALLS_ENABLED=true on the API and worker.