The Yap library

Dictation, speech to text and faster writing, explained with evidence.

Start with a topic, go down to a guide, a single question or a definition, and keep going: every number links to a fact page with the exact words of its source, and every source links to the original.

1,055

Questions

One question per page, answered in the first lines.

191

Glossary

The words of dictation and speech recognition, defined.

4,987

Facts

Every number we use, with the exact words of its source.

1,784

Sources

The studies, standards and official pages behind the facts.

Dictation

Dictation and Voice Typing: How to Talk Instead of Type

Dictation is speaking your words so a computer types them as you talk, and every current phone and computer does it for free with no voice training. Doing it well is a skill rather than a setting: compose each sentence before you say it, keep the microphone close, draft by voice and edit afterward, and proofread, because raw speech recognition still leaves errors. In a classic IBM study, novices learned to dictate letters as fast and as well as they wrote them within a few hours.

14 guides · 78 questions

Read the overview

Dictation software

Dictation Software: How to Choose the Right App

Start with the free dictation already built into your computer or phone, and pay only for what it cannot do: typing in every app, working offline, AI cleanup, a specialist vocabulary or full hands-free control. Four choices decide the rest (on-device or cloud, raw words or AI rewriting, subscription or one-time license, general or specialist), and a short test on your own writing settles the final pick better than any ranking.

10 guides · 71 questions

Read the overview

Typing and writing speed

Typing Speed: Average WPM and How to Type Faster

Most adults type roughly 35 to 55 words per minute on a computer keyboard and about 36 on a phone, while people speak at about 150 to 200 and read silently at about 240, so the hands, not the mind, usually set the pace of writing. Accurate practice and consistent finger use raise typing speed by tens of words per minute at most; the larger gains come from typing less, with text expansion, autocomplete and dictation.

14 guides · 81 questions

Read the overview

Dictation on Mac

Mac Dictation: How to Dictate and Control a Mac by Voice

Every Mac has free, built-in Dictation: turn it on once in System Settings > Keyboard, press the Microphone key or your shortcut in any text field and speak, and on Apple silicon Macs 48 of its 63 language variants can be processed on the Mac itself. Voice Control adds hands-free commands for the whole computer. When dictation fails, the cause is one of five things: the shortcut, the microphone, the language, Voice Control or the app you are typing in; when you need long, specialist or AI-formatted text, a dedicated Mac dictation app is the next step.

13 guides · 70 questions

Read the overview

Dictation on Windows

Dictation on Windows: Voice Typing, Voice Access, Apps

Every Windows 11 PC can dictate for free: press the Windows key + H to voice type into any text field, or turn on Voice Access to dictate and control the whole PC by voice, which has replaced the deprecated Windows Speech Recognition. Voice typing sends your audio to Microsoft's online speech service, while Voice Access and fluid dictation on Copilot+ PCs work on the device, so the right tool depends on whether you need privacy, offline use, accuracy on long text or hands-free control. A dedicated dictation app fills the gaps the built-in tools leave.

14 guides · 81 questions

Read the overview

Accessibility and health

Speech to Text for Accessibility, RSI and Disabilities

Dictation moves writing from the hands to the voice, so it can take most keyboard work off people with RSI, carpal tunnel syndrome, arthritis or a motor disability, and it lets people with dyslexia, dysgraphia or ADHD write at the speed they think. It is a tool rather than a treatment: it works best alongside medical care and good ergonomics, with the accommodations the law provides at school and work, and with recognition that handles the way you actually speak.

13 guides · 79 questions

Read the overview

Dictation for work and study

Dictation for Work: Doctors, Lawyers, Writers, Students

Dictation pays off in any job that is mostly writing: in controlled tests speech is about three times faster than keyboard entry, so a draft takes a third of the time and the saving goes into editing. What differs by profession is everything around the words: who checks the draft, which vocabulary it must handle and which confidentiality rules apply. That chain is being rebuilt as AI scribes and on-device recognition replace dictating to a typist.

13 guides · 83 questions

Read the overview

Dictation in your apps

Voice Typing in Google Docs, Word and Every Other App

Google Docs, Word, Outlook, PowerPoint, OneNote, ChatGPT, Claude and Claude Code have their own dictation button, but each one types only inside its own app and usually needs an internet connection or a paid plan. Gmail on a computer, Slack, terminals and Linux desktops have none, so the reliable setup is the built-in button where it is good enough plus one system-wide dictation tool that types into whatever text field has focus.

13 guides · 72 questions

Read the overview

Dictation on phones and tablets

Dictation and Voice to Text on iPhone and Android

Every iPhone, iPad, Android phone, Samsung Galaxy and Apple Watch already has free voice typing in its keyboard: turn it on once, tap the microphone key and speak, and in many languages the speech is turned into text on the device itself, even with no internet. Speaking is about three times faster than typing on a phone keyboard, and when voice typing fails the cause is often a setting, a permission, a swapped keyboard or a missing language download, not the microphone.

13 guides · 73 questions

Read the overview

Speech recognition

Speech Recognition: What It Is and How Well It Works

Speech recognition turns the sound of speech into text by converting audio into acoustic features and searching for the most probable words, and since about 2012 single neural networks trained on hundreds of thousands of hours of audio have replaced hand-built chains of models. On clean, read English the best systems now miss only a word or two in a hundred, but accuracy still depends on the audio, the speaker, the vocabulary and whether the model streams or runs on your own device, so word error rate measured on audio like yours is the number that matters.

13 guides · 73 questions

Read the overview

Transcription

Transcription: How to Turn Audio and Video Into Text

Transcription turns recorded speech into text, and for most recordings an AI transcriber now produces a usable draft in minutes, for cents per audio hour, on your own device or in the cloud. What you end up with depends on the audio, the transcription style and the review, so pay for human work, full verbatim, timestamps and speaker labels only where the use (legal records, research, captions, publication) demands them.

14 guides · 81 questions

Read the overview

Voice privacy and security

Speech to Text Privacy: Where Your Voice Goes

Dictation is only as private as the place your voice is processed and the rules of whoever processes it. On-device recognition keeps the audio on your hardware, while cloud recognition sends it to a server that may store it, train models on it or let people review it. Voice is personal data everywhere and biometric data when it identifies you, so privacy laws, HIPAA, legal ethics rules and recording consent statutes all apply, and the reliable way to dictate privately is on-device recognition with history, training and sharing settings switched off.

12 guides · 64 questions

Read the overview

Languages and accents

Dictation Languages, Accents and Voice Translation

Every major platform now dictates dozens of languages, but support, accuracy and features thin out quickly beyond English and a dozen large languages, so the right tool depends on the exact language, regional variant and accent you speak. Set the precise language and region, use a multilingual model when you mix languages or speak with a strong accent, and treat translation as a second step whose errors add to the recognizer's.

14 guides · 82 questions

Read the overview

Voice and AI

Speech to Text AI: Dictation, Assistants and Voice Agents

Every modern speech-to-text engine is AI, a neural network trained on large amounts of recorded speech, but it becomes generative AI only when a language model rewrites what it heard. That optional second pass (fillers out, punctuation and formatting in) is what the new AI dictation apps add, at a price in cloud processing, delay and the risk of changed words. Voice assistants and voice agents put the same recognizer in front of a model that acts and answers instead of typing, so choose by the job: your own words on the page, or a task done for you.

13 guides · 67 questions

Read the overview