Speech to text typed anywhere.
TypeTarget is a free browser extension that transcribes your audio directly into any text field you tell it to.
Free · MIT licensed · Chrome 116+
How it works
Pick a model and/or language.
Select an AI model to transcribe locally on your computer or let a cloud provider handle the compute for you. Together, they understand over 100 human languages.
Select an input.
Click the Start listening button in the TypeTarget panel, or right click on any page. Choose TypeTarget → Listen to this tab.

Choose an output.
Click the Start typing button in the TypeTarget panel or right click on any web page’s text field and Select TypeTarget → Type to this field.

Own your notes. Output and save to anywhere.
TypeTarget doesn’t store your notes. Instead, you tell it where you want it to write them to.
Notes you keep
Use any browser tab as the audio source.
- Lectures and courses
- Every word of a recorded or live class, straight into your study notes.
- Interviews and user research
- Transcribe a recorded session while you watch it back.
- Podcasts and social media
- Save a transcript of an entire video for show notes, a blog draft or a quote you need.
- Call notes into a CRM
- Sales or support calls, typed straight into the notes field of the record.
Captions
Read what a tab is saying, as it says it.
- Captions where there are none
- Streams, embedded players and old recordings that never shipped subtitles.
- Live webinars and conferences
- Keep the whole transcript, not just the slide you photographed.
- Meeting notes
- A web call in one tab, the transcript in your doc or notes app in the next.
- Deaf and hard-of-hearing access
- A running text version of any call or video, in a box you can scroll back through.
Into any AI chat
TypeTarget can type into the prompt box of any LLM web interface. Give the AI its instructions, and every message you send it gets handled the way you asked.
- Live roleplay
- Cast the AI as a character. What the other side says lands in its chat box as your lines.
- Translations
- Tell it to translate everything you send. Communicate with someone who speaks another language.
- Meeting copilot
- Ask for follow-up questions, action items or a plain-English version of the jargon.
- Language practice
- A tutor that corrects, explains or quizzes you on the audio you're listening to.
- Study aid
- Turn a lecture into flashcards, a summary or practice questions as it happens.
- Live commentary
- Annotate claims in a debate, a pitch or a keynote while it's still going.
You're my Spanish conversation partner. Everything I send is what someone just said in an English video. Reply in simple Spanish, as if you were answering them, then give me one word to learn from it.
Free forever.
No meter running, no cap on what you keep. Record as much as you want, for as long as you want.
- No recording limits.
- Leave it running through a full-day conference. TypeTarget puts no cap on how long a session can go.*
- No storage limits.
- TypeTarget doesn't store transcripts, so there's nothing to fill up. Every word goes straight into the field you chose, and you can save it as a .txt file on your own computer. No servers, no accounts, no logs.
- No account needed.
- Free and open source under the MIT licence. No trial, no paid tier, no account to create. The cloud hosted transcription models run on your own Google or Groq account, and both offer a free tier.
* TypeTarget sets no limit of its own. Hosted models (Gemini, Groq) apply their own free-tier quotas, which Google and Groq set and can change at any time. Local Whisper has no quota.
Frequently asked questions.
Does it transcribe my own voice?
Do I need an API key?
How do I get an API key?
Create API key button in the top right.- Name your key anything.
- Choose
Create new project, and name the project anything too. - Click
Create key. - Copy the key it gives you and paste it into the API key field in TypeTarget.
Groq. Go to console.groq.com/keys and click the Create API Key button.
- Give it a Display Name of anything you want.
- Copy your Secret Key and paste it into the Groq API key field in TypeTarget.
Why is my transcription slow?
Why is it transcribing the wrong language?
Model: larger models are better at telling languages apart. If you know the language being spoken, pick it under Spoken language instead of leaving it on Auto. You can also increase Chunk length, which gives the model more speech to recognize the language from.Where does my audio go?
What languages do you support?
Which browsers does it work in?
Can I load it without the Chrome Web Store?
typetarget. Open chrome://extensions by pasting it into the address bar, as web pages can’t link to it (in Edge it’s edge://extensions, in Brave brave://extensions). Turn on Developer mode in the top-right corner, click Load unpacked and select the extracted folder. To update later, download the new zip, replace the folder and press the reload arrow on TypeTarget’s card.