AI Voice Agent Cost-Per-Minute Calculator
Add up speech-to-text, the LLM turn and text-to-speech to see what a voice agent costs per minute, per call and per month.
An AI phone agent isn't one bill, it's three running at once. This ai voice agent cost per minute tool stacks them: speech-to-text to hear the caller, the LLM to work out the reply, and text-to-speech to speak it back. Enter your call length and volume, pick the model, and set the per-minute speech rates and how many LLM tokens a minute of talk uses. It shows the cost per minute, the voice bot cost per call, and the monthly total. Sizing a realtime api cost or comparing ai phone agent pricing across models? Change the model and watch the per-minute figure move. Everything runs in your browser against a dated price table.
Read the guide: AI Voice Agent Cost Per Minute: The Three BillsYour voice agent
Cost per minute
$0.0216
$0.1081 per 5-minute call
- Per call
- $0.1081
- Monthly total
- $108.15
- Minutes / month
- 5,000
Per minute
$0.0216
- Speech-to-text28%
- LLM3%
- Text-to-speech69%
A voice agent stacks three costs on every minute: speech-to-text to hear the caller, the LLM to decide what to say, and text-to-speech to speak back. Shave any one and the per-minute rate drops.
LLM prices updated January 2026. The STT and TTS rates are examples you can edit — real speech pricing varies by provider and voice quality. Telephony, streaming and per-session fees are separate. Confirm every rate against the provider's own pricing page before you budget.
How it works
- 1
Set call length and volume
Enter the average minutes per call and how many calls you handle a month. These turn the per-minute rate into a real monthly bill and a total minute count.
- 2
Pick the model and speech rates
Choose the LLM behind the agent, then set the speech-to-text and text-to-speech rates per minute. The defaults are examples — Whisper-class STT around $0.006 and neural TTS around $0.015 a minute — edit them to your provider's numbers.
- 3
Read the per-minute breakdown
The headline is the cost per minute; the donut splits it across the three layers so you can see which one dominates. The stats show the per-call cost, the monthly total and the minutes you'd serve.
Instant & 100% private — nothing is uploaded
Every calculation runs locally in your browser. The prompts, token counts and numbers you enter stay on your own device and are never sent to a server — nothing is stored, logged or shared.
Frequently asked questions
- How much does an AI voice agent cost per minute?
- It's the sum of three per-minute costs: speech-to-text, the LLM turn, and text-to-speech. With Whisper-class STT near $0.006, neural TTS near $0.015, and a small model handling the tokens, a minute often lands somewhere around two to four cents. Swap in your own model and speech rates to get your number.
- Why does a voice agent have three separate costs?
- Voice isn't native to a language model. The caller's speech has to be transcribed to text (STT), the text goes to the LLM to decide what to say, and the reply is turned back into audio (TTS). Each is a metered service with its own per-minute or per-token rate, so the true cost is all three added together.
- How many LLM tokens does a minute of conversation use?
- It depends on speaking pace and how much history you resend each turn, but a rough figure is 1,000 to 2,000 tokens a minute for a back-and-forth call. This tool splits that roughly 40% input and 60% output to price it — adjust the tokens-per-minute field to match your own transcripts.
- Are these figures exact?
- No. The LLM rates come from a shared table dated on the page, and the STT and TTS rates are editable examples that vary by provider and voice quality. Telephony minutes, streaming and per-session fees are separate line items this tool doesn't include. Treat the total as a close estimate and confirm each rate with the provider. The tool runs entirely in your browser and stores nothing.
Important
For planning and estimates only. Prices come from a published rate table dated on the page; providers change pricing without notice, and token counts here are approximations. Confirm against the provider’s own pricing before you budget or commit.
More tools
More from the Hivly network
Free sister tools on our other sites.