Dictate using your microphone and see your words appear as text instantly — powered by your browser's built-in speech recognition.
Google Chrome and Microsoft Edge have strong support for the Web Speech API's recognition feature. Firefox and Safari have limited or no support as of current versions.
Yes, your browser will ask for microphone access the first time you use this tool — required for speech recognition to work.
Speech processing is handled by your browser's built-in engine (Chrome's implementation uses Google's speech service). This tool itself does not separately upload or store your audio.
The underlying browser API supports many languages, though this tool is configured for English by default.
Browser speech recognition generally transcribes spoken words without inferring punctuation unless specifically spoken aloud (like saying "comma" or "period"), so adding punctuation manually after dictating is normal and expected.
Brief pauses are handled fine and recognition continues. A longer silence may cause the recognition session to end automatically, depending on your browser's specific timeout behavior.
Yes, the output appears in a standard editable text area, so you can correct any misheard words, add punctuation, or reformat the text directly before copying it.
This tool converts your spoken words into written text in real time, using the Web Speech API's SpeechRecognition feature built into supporting browsers. Click Start Listening, speak naturally, and watch your words appear as text — useful for quick note-taking, drafting messages, or hands-free writing.
Speech recognition support varies significantly across browsers. Google Chrome and Microsoft Edge (which is Chromium-based) have solid, actively maintained support for this feature. Firefox and Safari have historically had limited or no support for the SpeechRecognition API, so this tool works most reliably in Chrome or Edge. If your browser doesn't support speech recognition, this tool will let you know rather than failing silently.
When you click Start Listening, the tool requests microphone access (your browser will prompt for permission) and begins streaming audio to the browser's speech recognition engine. As you speak, recognized words are appended to the text area in real time. You can stop listening at any point, and the transcribed text remains in the box for you to copy, edit, or continue dictating further.
Speech recognition accuracy improves noticeably with a few simple habits: speaking at a clear, moderate pace rather than rushing, minimizing background noise where possible, and using a decent microphone rather than a laptop's built-in mic across a room. Recognition engines are generally trained on clear, well-enunciated speech, so mumbled or very fast dictation tends to produce more errors that need manual correction afterward.
Speech recognition in supporting browsers (like Chrome) is typically processed via the browser vendor's speech service, not by this website — this tool simply displays whatever text your browser's recognition engine returns. No audio or transcribed text is separately collected, logged, or stored by this tool.
These two tools are easy to mix up by name but work in opposite directions. Speech to text, covered here, listens to spoken words through your microphone and converts them into written text — useful for dictating rather than typing. Text to speech does the reverse, taking written text and reading it aloud using a synthetic voice — useful for proofreading, accessibility, or simply listening to content instead of reading it. If you have text and want to hear it, or you have speech and want to read it, Logrith's Text to Speech tool handles that opposite direction.
Many writers find that dictating a rough first draft, then editing afterward, produces faster and sometimes more natural-sounding writing than typing from scratch, simply because speaking tends to flow at a faster, more continuous pace than typing for most people. The transcribed text from dictation is rarely publication-ready as-is — it usually needs punctuation added and a pass for clarity — but having a rough draft to edit is often a faster starting point than facing a completely blank page, which is why dictation has become a popular drafting technique even among people who type quickly.
No speech recognition system is perfect, and certain words — technical jargon, uncommon names, homophones like "there" versus "their" — are more prone to misrecognition than common everyday vocabulary. Rather than stopping to fix every error as it appears, many people find it more efficient to dictate through the entire passage first, then do a dedicated editing pass afterward to catch and correct misheard words, since interrupting your flow of speech to fix small errors in real time tends to break the natural dictation rhythm that makes this method faster than typing in the first place.
Speech recognition engines are generally trained on broad, everyday language, which means specialized vocabulary — technical jargon, brand names, uncommon proper nouns — gets misrecognized more often than common words. When dictating content heavy with specialized terms, speaking slightly more slowly and clearly around those specific words, and planning for a thorough proofreading pass focused specifically on technical terminology, tends to produce better final results than expecting the recognition engine to catch unfamiliar vocabulary correctly on the first attempt without any extra care.