How can we help?

Chrome remembers your answer to its microphone prompt. If you chose Block, or Saypen says “Microphone access is blocked”, allow it again:

1. Open Saypen’s setup page (right-click the Saypen icon › Options) and click Open microphone settings, or go to chrome://settings/content/microphone.
2. Under “Not allowed to use your microphone”, find the entry whose address starts with chrome-extension://. That entry is Saypen’s setup page.
3. Open it and set Microphone to Allow.
4. Go back to the setup page and reload it. Step 1 should now say Allowed.

When Chrome asks again, choose “Allow while visiting the site”. Saypen then uses the microphone only while voice typing is on.

Still blocked? At the top of chrome://settings/content/microphone, check that the right microphone is selected, and check that your computer’s own privacy settings allow Chrome to use the microphone.

Saypen can only use the languages that Chrome can recognize on your computer, and the setup page lists exactly those. On Chrome 156 for Windows that was 24 language variants, about 16 languages, for example English, Spanish, French, German and Japanese.

Albanian, Dutch and Ukrainian are not available in Chrome yet. We can’t add a language ourselves: when Chrome adds it, it appears in Saypen’s list. Meanwhile you can dictate in another language you speak.

To recognize speech on your computer, Chrome needs a speech model for each language, about 200 MB on disk. When the setup page says “Needs a one-time download”, click Download. Chrome, not Saypen, downloads it from Google. It usually takes less than a minute, and the status then says Ready. Works offline.

Chrome keeps the model for later use and shares it with all websites, so you download each language only once.

Check your internet connection and click Download again. The model comes from Google, so a network that blocks Google’s download servers can stop it. Make sure your disk has a few hundred megabytes free.

Saypen needs Chrome 139 or newer on a desktop computer, with on-device speech recognition available. Update Chrome (Menu › Help › About Google Chrome), restart it, then open Saypen’s setup page again. Saypen doesn’t work in Chrome on phones or tablets.

Saypen is made for Chrome on desktop computers. We have tested it on Windows so far. If on-device speech recognition isn’t available in your Chrome, the setup page tells you. We’d be glad to hear how it works for you.

Chrome doesn’t let extensions run on its own pages (addresses starting with chrome://), on the Chrome Web Store or in its PDF viewer. There the Saypen toolbar button shows a short explanation instead of listening. Use voice typing on normal websites.

No text field was selected when the phrase was finished, so Saypen kept the text in its bar instead of losing it. Click Copy to put it on the clipboard and paste it where you want, or Close. Next time, click into the field first, then press Alt+Shift+D.

Many sites don’t use a plain text box but a rich editor. Saypen types into text inputs, text areas, plain editable fields and editors built on ProseMirror, Slate, Lexical, Quill, Draft.js, CKEditor 5, TinyMCE, CodeMirror 6 and Monaco. These are the editors behind many mail, chat, AI-chat and social sites, and all 12 kinds were tested on our own test pages.

Google Docs uses its own kind of editor, and we haven’t tested it yet.

If text doesn’t appear in an editor, click into it again and retry. If it still doesn’t work, write to us with the site’s address. Meanwhile, you can dictate in the “Try it here” box on the setup page and copy the text from there.

On purpose. While you speak, Saypen’s bar shows the words it hears, so a spoken password would be visible on screen. Saypen never types into password fields. Type passwords yourself.

No. In chat apps, “new line” makes a line break and doesn’t send the message, and “new paragraph” starts a new paragraph. Send the message yourself when you’re ready.

“delete that” removes the last phrase Saypen typed, and only while it is still right before the cursor. If you have edited or moved away from it, Saypen leaves your text alone and shows “Couldn’t delete it: the text has changed”.

Saypen stops on its own:

• after a silence: 1 minute by default. On the setup page you can choose 30 seconds, 2 minutes, 5 minutes or never;
• when you switch to another tab, or to another window or app (after about 1.5 seconds);
• when the page navigates, or when the tab closes;
• when you say “stop listening”.

If the bar says “The microphone was disconnected”, check your headset or microphone and start again.

The default is Alt+Shift+D. Open chrome://extensions/shortcuts (or click Change shortcut on the setup page) and set a new key for “Start or stop voice typing”. If another extension already uses those keys, pick different ones. The Saypen button in the toolbar always works too: pin it from the puzzle-piece menu.

Open the setup page (right-click the Saypen icon › Options). Under Settings you can turn off automatic punctuation and capital letters, voice commands, and the short sounds when voice typing starts and stops, and choose the silence timeout. Settings are stored only on this computer.

Only if you allow it. Open chrome://extensions, click Details under Saypen and turn on Allow in Incognito. The same privacy rules apply there.

Saypen uses Chrome’s on-device speech model. It’s good for dictation, but not perfect, and names and specialist words can come out wrong. These help:

• Choose the language and the variant closest to how you speak, for example the right English for your accent, if the list offers several.
• Speak at a normal pace, in whole sentences, with a short pause between them.
• Use a headset or a microphone close to you, in a quiet room.
• Watch the bar: it shows what Saypen hears before the phrase is typed.
• Say “delete that” to remove the last phrase and say it again, or fix a word with the keyboard.
• Test your microphone in the “Try it here” box on the setup page.

The bar on the page says “Listening”, Saypen’s toolbar icon shows a red “ON” badge. Your computer may also show that the microphone is in use. The microphone is open only while voice typing is on, or while the “Try it here” box on the setup page is listening.

No. Chrome turns your voice into text on your computer, and Saypen itself makes no network requests. It never records, saves or sends audio and keeps no history. The only download is the speech model, which Chrome fetches from Google when you click Download. Details are in the privacy policy.

Right-click the Saypen icon in the toolbar and choose Remove from Chrome, or open chrome://extensions and click Remove under Saypen. This also removes Saypen’s settings.

The speech models belong to Chrome, which manages them and shares them with all websites. If Saypen is still listed at chrome://settings/content/microphone (an address starting with chrome-extension://), you can remove that entry too.