Speech to text that types as you talk

Phone numbers, dates, and email addresses come out right the first time, even with background noise, from the model ranked #1 for streaming speech-to-text accuracy. Free, no sign-up.
Start buildingContact sales
Click or press Space to start a transcription
Tap to start a transcription
Need a topic?
If you could teleport anywhere right now, where would you go?

Join the teams making the switch to Cartesia

Artificial Analysis

Ranked #1

in Speech Arena leaderboard & Speech to Text leaderboard by Artificial Analysis

How to convert speech to text

01

Press the microphone button and allow access when your browser asks.

02

Talk the way you normally would. Each word appears as you say it, and the transcript marks where you start and stop speaking.

03

Select the transcript, copy it, and paste it into your document, email, or notes.

Accurate on the numbers and emails other models get wrong

This tool runs on Ink-2, Cartesia's streaming speech-to-text model. Ink-2 ranks first on the VoiceArena US-English leaderboard and Artificial Analysis' streaming benchmark, and it writes out phone numbers, emails, and UUIDs correctly the first time. It splits the transcript into turns on its own and stays accurate in background noise without a separate cleanup filter.

Good for

  • Voice agents
  • Live captions
  • Dictation
  • Meeting notes
  • Call analytics

Add speech to text to your app

Stream live audio over a WebSocket to the same Ink-2 model this tool uses, with the Cartesia API and SDKs. Keyterm prompting biases recognition toward the names and jargon your product hears most. For recordings, the batch API takes a whole file and returns one transcript. Enterprise plans add on-prem, VPC, and OEM deployments with zero data retention.

Get an API keyRead the docs

Explore more voice tools

All free tools

AI voice generator

Type a script and hear it read by a lifelike AI voice. Free to try, no sign-up.

Explore

Realistic text to speech

Speech that sounds recorded, from the model ranked first in blind listening tests on Artificial Analysis.

Explore

AI voiceover generator

Realistic narration from any script, for videos, courses, and podcasts.

Explore

Speech to text FAQs

Is there a free speech to text tool?

Yes. The tool on this page turns your speech into text for free, with no account, up to a daily limit. A free Cartesia account's monthly credits cover up to 111 minutes of transcription through the API, and paid plans add more.

How accurate is speech to text?

Ink-2, the model behind this tool, is ranked #1 for streaming speech-to-text accuracy. It writes phone numbers, emails, and UUIDs correctly and stays accurate with background noise.

What languages does Cartesia speech to text support?

Ink-2 transcribes English, Spanish, French, Hindi, and Japanese. The tool on this page listens in the page's language when it is one of those five, and in English otherwise. The API takes all five.

Can I convert an audio file to text?

The tool on this page transcribes what you say into your microphone, live. To transcribe a recording, send the file to the batch speech-to-text API, which returns one transcript for the whole file.

What is speech to text?

Speech to text (also called STT, ASR, or transcription) turns spoken audio into written words. A streaming model like Ink-2 writes the transcript while you are still talking instead of after you stop.

What is turn detection in speech to text?

Turn detection decides when a speaker is done, which is what keeps a voice agent from interrupting or leaving dead air. Most stacks add a separate voice-activity model for this. Ink-2 emits turn events like turn.eager_end and turn.end itself, which means one fewer model in the stack and faster responses.

Can I run Ink-2 in my own infrastructure?

Yes. Ink-2 deploys on-prem (including air-gapped environments), in your own VPC on AWS, GCP, or Azure, or as an OEM license embedded in your product. These deployments are available under enterprise contracts.