Dictate in any app · AI chat
Text-to-speech needs a script, and the script is the part you have to write. Typally types what you say straight into ElevenLabs — hold a key, talk, let go — so you can dictate anywhere ElevenLabs lets you type.
This is a vendor guide from the makers of Typally. Anything we say about ElevenLabs is worth verifying — third-party features change without notice.Text-to-speech is the product — it reads your text aloud. That is the opposite direction from dictation.
People use ElevenLabs to turn scripts into narration, which means someone typed the script first. Writing a natural-sounding script by typing is oddly hard, because typed prose does not sound like speech.
In practice people use it for: Narration scripts, voiceover copy, audiobook text, and the dialogue in a generated scene.
Nothing changes inside ElevenLabs itself. Typally sits in the tray and types into the focused window:
Install, sign in, and it waits in the tray.
The cursor position is the whole instruction — ElevenLabs routes the keystrokes from there.
Push to talk, say your piece, release to transcribe.
Dictation and text-to-speech are opposite processes. Do not expect the platform to transcribe your voice — it generates one.
Typally works in the opposite direction to ElevenLabs and pairs well with it. Speak the script, let Typally clean it into text, then feed that text in. Prose that started as speech has the rhythm of speech, which is exactly what makes a synthetic read sound natural rather than stilted.
A creator whose narration kept sounding stilted starts speaking the script first and pasting the transcript in. The synthetic read improves immediately, because the sentences were built for a mouth rather than a page.
Dictate the script, then have it read aloud. Text written by speaking already has the rhythm of speech, which is exactly what a synthetic voice needs to sound natural.
Dictation and text-to-speech are opposite processes. Do not expect the platform to transcribe your voice — it generates one.
Its core product generates speech from text. Dictation is the reverse: your voice becoming text you can then feed in.
Because spoken-origin text sounds natural when read aloud, and typed prose often does not.
Yes — it is an ordinary text area.
Free plan: 100 minutes of dictation a month, no card. Pro is $7.99/month if you want more.
Also dictate in: Midjourney · Runway · Suno · Grok · Character.AI · see all apps