Voice agent cost calculator
Price an AI phone or voice agent per minute and per month: speech-to-text, the model and the voice.
Free, runs in your browser, no sign-up. Updated 2026-10-25.
Result
$0.032 per minute ($0.13 per call, $316 for 10,000 minutes a month). Telephony not included: enter your carrier's rate.
Telephony is $0 until you enter your carrier's per-minute price.
- Speech-to-text
- $0.0077
- LLM
- $0.0089
- Text-to-speech
- $0.015
- Telephony
- $0 (enter yours)
| Part | Source |
|---|---|
| Nova-3 Monolingual (streaming)Regular pay-as-you-go streaming price. Deepgram showed a limited-time promotional price of $0.0048/min on 2026-10-25. | Checked 2026-10-25 |
| Claude Sonnet 5.5 | Checked 2026-10-23 |
| Aura-2Pay-as-you-go. | Checked 2026-10-25 |
Assumptions
Every number the result depends on, and where it comes from. Change any value and the result updates.
| Assumption | Value | Source |
|---|---|---|
| Call minutes per month | minutes | OceanLab planning assumption. Use your own call volume.Reviewed 2026-10-25 |
| Share of the call sent to speech-to-text | % | OceanLab planning assumption: streaming STT usually listens for the whole call. Lower it if you only stream while the caller speaks.Reviewed 2026-10-25 |
| Share of the call the agent is speaking | % | OceanLab planning assumption: an even conversation. Support calls where the agent explains a lot run higher.Reviewed 2026-10-25 |
| Characters of speech per minute of audio | chars | OceanLab planning assumption taken from ElevenLabs' API pricing page, which equates 1K characters with about 1 minute of audio. Edit it for your voice speed.Reviewed 2026-10-25 |
| Agent turns (LLM calls) per minute | turns | OceanLab planning assumption: a reply every 15 seconds. Check your transcripts.Reviewed 2026-10-25 |
| System prompt + tools (tokens per LLM call) | tokens | OceanLab planning assumption. Paste your prompt into the token counter.Reviewed 2026-10-25 |
| Conversation history added per turn (tokens) | tokens | OceanLab planning assumption: a transcribed caller turn plus the agent reply.Reviewed 2026-10-25 |
| Average call length (minutes) | minutes | OceanLab planning assumption. Used for history growth within a call.Reviewed 2026-10-25 |
| Agent reply (tokens) | tokens | OceanLab planning assumption: one or two spoken sentences.Reviewed 2026-10-25 |
| Share of the prompt served from cache | % | OceanLab planning assumption: the prompt repeats on every turn. Set 0 if you do not use caching.Reviewed 2026-10-25 |
| Telephony / SIP per minute (USD) | USD | Enter your vendor's price. Not verified for any carrier, so it starts at 0. Enter your provider's rate.Reviewed 2026-10-25 |
Ask OceanLab about your voice agent
Opens an email to OceanLab with this result and a link to your inputs. Nothing is sent until you press send, and there is no sign-up.
How this is calculated
Per minute = speech-to-text on the caller audio + LLM calls for each agent turn + text-to-speech for the agent's share of the call + telephony.
Per-character voice prices are turned into minutes with an editable characters-per-minute assumption (the default follows ElevenLabs' pricing page: about 1,000 characters per minute).
Telephony rates were not verified for any carrier, so they start at 0. Enter your provider's per-minute price.
Sources
- OpenAI speech pricing · checked 2026-10-25
- Deepgram speech pricing · checked 2026-10-25
- ElevenLabs speech pricing · checked 2026-10-25
- Google speech pricing · checked 2026-10-25
- Anthropic API pricing · checked 2026-10-23
- OpenAI API pricing · checked 2026-10-23
- Google API pricing · checked 2026-10-23
- DeepSeek API pricing · checked 2026-10-23
- MiniMax API pricing · checked 2026-10-23
Questions about this tool
What is included per minute?
Speech-to-text on the caller audio, the LLM calls for each agent turn, and text-to-speech for the agent's share of the call. Telephony is a separate input.
Why is telephony at zero?
No carrier's rates were verified, so it starts at 0. Enter your provider's per-minute price.
How are per-character voice prices turned into minutes?
With an editable characters-per-minute assumption. The default follows ElevenLabs' pricing page, which equates about 1,000 characters with 1 minute.
Where do the prices come from?
Each price is read from the provider's official pricing page and shows the date it was checked. Prices that could not be verified are left out, or start at 0 for you to fill in.
Other free tools
OceanLab product builds like this
Products OceanLab designed and built. Each case study covers the screens, the scope and the rules it was built for.
Pitline - AI Voice Agent and Call Review Console for a Garage Chain
Designed and built a clickable staff console for an AI phone agent that books, moves and cancels car servicing and MOTs, reads out job status and answers common questions for a chain of 14 Midlands garages - with every call reviewable turn by turn, every tool call on the record and a warm transfer to a person whenever the agent should not decide.
Cayuco - Voice AI Contact Centre for Health Insurance Member Services
Designed and built a clickable cloud contact centre for a Panama City health insurer's member services line: an AI voice agent that says it is automated, verifies callers and handles claim status and card replacement, an agent desktop with live transcript, suggested answers, secure card payments and AI call summaries, and review tools for every AI-handled call.
Pahana - Trilingual AI Support Agent with Safe Human Handoff
Designed and built the staff console for a trilingual AI support agent at a Colombo telecom operator: a dashboard of containment and satisfaction, every conversation with its tool calls and guardrail events, a live monitor where team leads take over a chat, and a handoff queue that makes sure no customer passed to a person is lost.
