Text to Speech
Paste your text and press Play. By default only voices that run on your device are used, so your text stays with you.
Voice and speed
Online voices are generated by your browser maker's servers: when you pick one, your browser sends the text to that company. Talisman never sees it, but it does leave your device.
No voices found on this device.
Which voices does it use?
Talisman uses the speech voices that come with your operating system and browser through the standard Web Speech API. The list depends on your device: Windows, macOS, iOS, Android and ChromeOS include several voices, and you can usually download more in the system speech or accessibility settings.
Does my text leave my device?
Not with on-device voices, which are the default. Some browsers also offer online voices (for example Chrome's "Google" voices or Edge's "Online (Natural)" voices). Those are produced on the browser maker's servers, so the browser sends them your text. Talisman lists them separately and only after you switch them on.
Can I download the audio as an MP3?
Not yet. Browsers play Web Speech voices directly to the speakers and do not give web pages access to the audio, so it cannot be saved. A downloadable version needs a speech model that runs inside the page, which we plan to add later.
Why is long text read sentence by sentence?
Some browsers stop speaking after about 15 seconds of continuous speech. Talisman splits your text into sentences (and long sentences at commas) and queues them, so long articles play to the end and the current sentence can be highlighted.