Speech to Text. Your words, appearing live.
Honest note: Digital Heroes never sees or stores your audio or text, but speech recognition itself is done by your browser/OS built-in speech service (e.g. Chrome may process audio on Google’s servers). If you need fully offline dictation, use your OS’s native dictation feature.
Transcript (editable)
0 words · 0 charactersTips for better accuracy
Speak clearly at a normal pace, pick the exact locale you speak (e.g. English, India vs English, US), and use a headset mic if you can. Recognition pauses on silence, with auto-restart on, it picks back up automatically so long dictation sessions keep flowing.
How to use the Speech to Text tool
This tool turns what you say into text with the speech recognition built into your browser, the Web Speech API. Words appear in a gray preview while you speak and move into an editable transcript once the browser settles on them. It is for dictating notes, drafts, messages and meeting logs, and it works best in Chrome or Edge.
Step by step
-
Use a supported browser
Open the page in a recent Chrome or Edge on desktop or Android. If your browser has no speech recognition, the tool is replaced by a notice saying so.
-
Pick your language
Choose the exact locale you speak under Language, such as English, India or English, US. The list has 38 options and starts on your browser's language when it is listed, otherwise on English, US.
-
Choose how results are joined
Under Append each result as, pick Continuous text (spaces), New line per result or Blank line per result. A result is usually one phrase between pauses.
-
Set the options
Continuous listening, Show interim results, Auto-restart if it cuts out and Auto-capitalize sentences are on by default. Tick Timestamp each result to start each result with the time since you pressed start, such as [01:24].
-
Dictate
Press Start dictating and allow the microphone when the browser asks. The status changes to listening, the timer runs and the level meter moves as you speak. Press Stop when you are done.
-
Edit and save
Fix anything in the Transcript box by typing. Copy text copies it all, Download .txt saves a file named with the date and time, and Clear empties the box.
Reading the result
- The status badge reads idle, starting…, listening or restarting…. Restarting means the browser ended its session, for example after a silence, and Auto-restart is opening a new one.
- Gray italic text above the transcript is the interim guess. It can change as you keep speaking; only final results are added to the transcript.
- The counter above the box shows words and characters for the whole transcript, including anything you typed or edited yourself.
- Auto-capitalize sentences uppercases the first letter of each result and any letter after a period, question mark or exclamation mark inside it.
Tips and common mistakes
- Language, Continuous listening and Show interim results are locked while you dictate. Press Stop, change them, then start again.
- To capture a single phrase and stop, untick both Continuous listening and Auto-restart if it cuts out. With auto restart on, listening resumes after every result.
- Speak at a steady pace, use a headset microphone if you can, and pick the regional locale you actually speak, as the tips card on the page suggests.
- If you see Microphone permission denied, allow the microphone for this site in your browser settings and press Start dictating again.
- The transcript is not saved. Copy or download it before you close or reload the page.
Privacy and limits
This site never receives your audio or text. Recognition is done by your browser's speech service, and in Chrome that service may process the audio on Google's servers, so dictation usually needs a connection. The transcript stays in the page and is not stored, and the microphone is released when you press Stop.
Everything a dictation tool should do.
- Live voice-to-text transcription in 30+ languages and locales, with real-time interim results as you speak.
- Continuous dictation with auto-restart, so long sessions keep flowing even when recognition pauses on silence.
- Timestamp each result to build time-coded notes or meeting logs, and auto-capitalize sentences for you.
- Edit the transcript freely in the text area before saving, with a live word and character count.
- Copy or download your finished transcript as a plain .txt file with one click.
- Private by construction: your text is never uploaded to Digital Heroes, and no sign-up is needed.
Frequently asked questions.
Is Speech to Text free?
Is it private? Is my audio uploaded anywhere?
Does it work offline?
Do I need to install anything or create an account?
Which browsers support this tool?
Published · Last updated .
