Gujarati Voice Typing

If you proofread one thing in dictated Gujarati, make it the anusvara — ં. It marks nasalisation that is light in ordinary speech, sometimes barely audible, and a recogniser working from sound alone drops it more often than it keeps it.

In writing it is not optional. It distinguishes words, and a sentence full of correctly spelled but un-nasalised forms reads as careless rather than as a transcription artefact. Because the omission is systematic rather than random, scanning specifically for missing anusvaras is far quicker than rereading the whole text.

Everyday Gujarati mixes English constantly — around business, around family arrangements, around ordinary objects. The recogniser transcribes those words in Latin script, so you get appointment and booking sitting inside an otherwise Gujarati sentence rather than nasalised Gujarati spellings of them.

For a WhatsApp message that is probably what you would have written anyway. For anything printed, it is a deliberate conversion pass — and the transliteration tool is faster for it than retyping.

Gujarati is written without the shirorekha, the horizontal line that runs across the top of Devanagari. It is a script fact rather than a dictation one, but it changes proofreading: without that line tying a word together, a wrong letter in the middle is visually less obvious than the same error would be in Hindi or Marathi.

Reading dictated Gujarati a little more slowly than feels necessary is the practical consequence.

The microphone is your browser’s. On Chrome and Edge the audio is sent to the browser vendor’s speech service and text comes back; WorldTyping is never in that path and never receives audio. The text stays in your browser, saved locally so a reload does not lose your draft.

That is also why dictation needs an internet connection, and why a browser without Gujarati speech support will not respond to the button.