Today’s tool: the ElevenLabs API. It’s the reason a daily video goes out in my voice on mornings when I haven’t said a word.
Every evening one of my agents writes a short script, sends it to ElevenLabs, and gets back an MP3 of a cloned “Jon Jones” voice reading it. The rest of the pipeline adds captions, visuals and music, then schedules the video. I don’t record anything. Here’s how the ElevenLabs API piece works and what it actually costs.
The whole integration is one POST
This is the real call from my video pipeline, trimmed:
POST https://api.elevenlabs.io/v1/text-to-speech/{voice_id}
xi-api-key: $ELEVENLABS_API
Accept: audio/mpeg
{"text": "...script...",
"model_id": "eleven_multilingual_v2",
"voice_settings": {"stability": 0.55, "similarity_boost": 0.75, "style": 0.3}}The response body is the MP3, with no job ID and no polling. The agent writes the bytes to a file and moves on to the next step. If your agent can make an HTTP request, it can talk.
Pin model_id explicitly. A pinned model means the voice can’t change under you overnight, and you’ll know exactly which line to edit when that model gets retired. (I wrote up how I handle model retirements here.)
The receipt: what 30 days actually used
I pulled this from the API’s own /v1/history and /v1/user/subscription endpoints this morning:
⚡ GET THE AI EDGE
Weekly AI tips that actually save you time and money. No fluff, no hype — just what works.
- 37 generations in my cloned voice from Oct 5 to Nov 4, on 28 different days. Together they used 18,668 characters, and the median script was 468 characters.
- Billing is literal. For all 34 text-to-speech calls, the credits charged matched the script length exactly, one credit per character. The other 3 were speech-to-speech tests, which bill by audio instead, at 1,385–1,435 credits each. That’s about 2.6x what a text script of similar length costs.
- One account covers the whole fleet. In the same 30 days, 237 generations across 13 voices for different brands used 95,789 characters on one Creator-tier subscription with a $22.00 monthly subtotal. This cycle has used 24,482 of 216,358 characters so far.
So one daily narration costs pennies, and the plan has plenty of headroom. Characters are the meter, so trim the script before you send it, not the audio after.
The gotcha: big numbers get read as years
This is the one rule my pipeline had to learn. ElevenLabs reads any 4-digit number like a year. “2072 adults” comes out as “twenty seventy-two adults.”
The fix lives in the script-writing prompt, not in ElevenLabs. My video agent’s instructions now include this rule: any number with four or more digits in the spoken text must be spelled out, so “1,200 sessions” becomes “twelve hundred sessions.” Real years stay as digits because they read correctly. One more rule: always put a space between a number and its unit, because “22months” gets mangled.
Text on screen is a different case. Titles can keep their digits, because nobody reads them aloud.
Three tips if you’re wiring it into an agent
- Read
/v1/user/subscriptionbefore a big batch. It returnscharacter_count,character_limitand the reset date. Unlike some APIs I use, you never have to guess what’s left. - Use
/v1/historyas your audit log. It keeps the exact text of every generation, so you can always check what your agent actually said in your voice. - Make one render feed several platforms. My single narration becomes a TikTok, an Instagram Reel and a YouTube Short. (Build once, ship everywhere.)
The takeaway: a voice API turns “I should make videos” into a step your agent runs on its own every day. Keep the script short, spell out the big numbers, and check the meter before a batch.
Want your own voice in a pipeline that runs without you? Book an automation strategy session.
📥 FREE: THE AI PLAYBOOK
The exact tools and workflows I use to run a one-person agency. 25 years of marketing experience distilled into an actionable guide. Yours free.
AI & Digital Marketing Strategist
With over 25 years in digital marketing, e-commerce, sales, and international business, Jon Jones is a seasoned entrepreneur dedicated to empowering small businesses through innovative AI solutions. His journey began in investment banking on Wall Street, guiding companies through capital acquisition and complex markets, and has expanded to include expertise in web development, brand design, product development, and courses and training.
Fluent in Mandarin and with a rich foundation in both finance and tech, Jon launched JonJones.ai to provide accessible, transformative tools for small businesses to streamline operations, generate leads, and build sustainable growth. His mastery of AI, automation, and digital strategy has enabled numerous businesses to bridge the gap between traditional practices and modern technology, ensuring they thrive in today’s digital landscape.
