Type a word or phrase and hear it spoken aloud in multiple languages using the Speech Synthesis API.
Type a word or phrase, pick a language, and the page speaks it aloud through the browser's Speech Synthesis API, using voices your operating system already provides. Synthesis is local: nothing is recorded, uploaded, or stored. The tool plays text; it does not judge your speech, so 'checking' means listening and comparing yourself. The surprise for most users is that the voice is a machine voice owned by their device: the same word sounds different from browser to browser, and a language with no installed voice silently falls back to an unrelated accent.
A concrete input and expected output from the current implementation.
Input
Bonjour, comment ça va ? (with French selected)
Expected output
The output is audio: the device's French voice reads the phrase with standard French pronunciation - bonjour as /bɔ̃.ʒuʁ/ and comment ça va as /kɔ.mɑ̃ sa va/. There is no transcript or score to read back; the result is what you hear.
French spelling maps to sound almost rule-for-rule, so any French voice renders the phrase with the same phonemes; only timbre and accent shading vary. The tool converts the typed text to speech rather than looking up a recording, which is why the sound depends on the installed voice.
The tool lists the voices your browser and operating system provide, each tagged with a language code such as en-US or fr-FR. The list loads at runtime and can take a second to appear, and it differs from machine to machine, so the French voice on a Mac is not the one an Android phone uses.
The text is handed to the speech engine with the selected language's rules, and the engine renders it locally through your speakers; nothing is sent to a server. Many browsers refuse to start audio until the page has received at least one click or tap, and some engines cut off or skip long passages, so short phrases are the reliable input.
Most browsers hold back audio until the page has had at least one click or tap, so click anywhere on the page first, then play again. Also confirm a voice exists for the chosen language, and keep the input short, since some engines stutter or stop on long paragraphs.
No. It plays a machine rendering of the text; it does not listen to you, score your speech, or compare recordings. Treat it as a reference to imitate by ear, and for anything important, defer to a native speaker or an audio dictionary.
The example and behavioral notes were checked against the browser implementation. Standards and primary references below define the relevant format, formula, or platform behavior.
Interactive periodic table - click elements, enter quiz mode, and guess elements from their properties. Gamified learning.
Explore the periodic table with element details, properties, and category filtering.
Browse and search ASCII, Unicode, and binary character codes with copy-to-clipboard.
Test your knowledge of semantic HTML elements with quizzes and best-practice tips.
Structured typing lessons with finger placement guide. Track WPM improvement over time with localStorage persistence.
Timed multiplication drills with accuracy tracking per table (1-12). Progress chart saved across sessions.