Speech to text in English has come a long way from the clunky, pause-and-speak systems of the early 2000s. Today's technology understands natural speech at near-human accuracy levels, handles accents and regional dialects, and can transcribe at speeds far beyond what any typist can achieve. Yet millions of people still default to the keyboard for everything from emails to long-form documents, simply because they have not experienced modern voice dictation firsthand.
This guide covers everything you need to know about using speech to text for English — how the technology works, what separates a great dictation experience from a mediocre one, and practical techniques for getting the most out of your voice.
How Modern Speech to Text Works
Modern speech recognition systems analyze the audio waveform of your voice, break it into phonemes, and then use statistical language models to determine the most likely word sequence. What makes today's systems dramatically better than earlier generations is the size and quality of the training data. Contemporary speech recognition models have been trained on hundreds of thousands of hours of English speech across accents, recording environments, speaking styles, and domains.
The result is accuracy that rivals human transcription for clear speech. Under good conditions — a quiet room, a reasonable microphone, and a native or near-native English speaker — modern systems regularly achieve error rates below 5 percent. For everyday dictation tasks like composing emails, writing documents, or filling out forms, that level of accuracy means you can speak naturally and expect to see what you said on screen.
English Dialects and Accent Support
One of the most common concerns people have about speech to text is whether it will understand their accent. The short answer is that modern systems handle a wide range of English accents quite well. American English, British English, Australian English, Indian English, and many others are all supported with high accuracy by leading voice recognition tools.
The longer answer is that accuracy does vary by accent, and accents that are less well represented in training data tend to see slightly higher error rates. If you speak with a regional dialect — Scottish English, Caribbean English, African American Vernacular English — you may find that some words are occasionally misrecognized. The practical remedy is simple: speak at a natural but deliberate pace, and use the custom vocabulary feature available in most dictation apps to add words or phrases that are commonly misheard.
The Speed Advantage
The most compelling reason to adopt speech to text is the speed gap between speaking and typing. The average person types between 40 and 60 words per minute. The same person speaks between 120 and 180 words per minute. That is a two-to-four times productivity multiplier for any task that involves producing text.
The gap is even wider once you factor in thinking time. When you type, you tend to think in short bursts — compose a phrase, type it, pause, compose the next phrase, type it. When you dictate, you can think in longer arcs because the bottleneck of physical typing is removed. Many people find that their first drafts are not only faster but better when dictated, because they capture the natural flow of their thinking without interruption.
What to Look For in a Speech to Text App
Not all speech to text software is equal, and the differences matter for everyday use. Here are the key factors to evaluate:
Latency
Latency is the delay between when you finish speaking and when text appears on screen. High latency — anything over two to three seconds — destroys the feeling of dictation as a natural activity and forces you to wait before you can see and correct what you said. The best tools deliver text in under a second, making the experience feel instantaneous.
System-Level Integration
Some dictation apps only work in their own interface, requiring you to transcribe text there and then copy it to wherever you actually need it. System-level dictation tools insert text directly at your cursor position in any application — email, documents, chat, web forms, code editors. This is the difference between a tool you actually use and one you abandon after the first week.
Accuracy on Specialized Vocabulary
General English accuracy is table stakes. The real differentiator is how well a system handles professional terminology — medical terms, legal language, technical jargon, brand names, and industry-specific vocabulary. Tools that let you add custom vocabulary to the recognition engine are essential for professional use.
Privacy
Your dictated speech may contain sensitive information — client names, medical details, financial data, private thoughts. It matters where that audio goes and how it is handled. Look for tools that are transparent about their data practices and do not store audio recordings unnecessarily.
Getting the Best Accuracy from English Dictation
Even the best speech recognition system needs a little help from its user. These techniques will significantly improve your results:
Use a Quality Microphone
The built-in microphone on most laptops is adequate for casual dictation in a quiet room. For professional use or noisy environments, a dedicated headset microphone makes a noticeable difference in accuracy. The microphone does not need to be expensive — any USB headset in the $30 to $60 range will outperform most built-in mics for dictation purposes.
Speak in Complete Phrases
Dictation accuracy improves when you speak in complete, grammatically coherent phrases rather than isolated words. The language model uses context to disambiguate words that sound alike, so "I need to write a letter to the board" will be transcribed more accurately than "I... need... to write... a letter... to the... board" with long pauses between each word.
Include Punctuation Commands
Most dictation systems recognize spoken punctuation commands like "comma," "period," "new paragraph," and "question mark." Using these while you dictate produces cleaner text and reduces editing time afterward. It takes a few sessions to make punctuation commands feel natural, but once you internalize the habit, your dictated text requires minimal cleanup.
Edit After, Not During
The urge to stop and correct every small error as it happens is one of the biggest productivity killers in dictation. Speak your full draft first, then go back and fix errors with the keyboard. You will be surprised how few corrections are needed when you review the complete text.
The Best Use Cases for English Speech to Text
Speech to text delivers the biggest productivity gains in tasks that involve composing substantial amounts of text. Email is the most universally applicable use case — the average knowledge worker sends dozens of emails per day, and dictating responses rather than typing them can recover hours of time every week. Long-form writing is another strong use case: blog posts, reports, research notes, and meeting summaries all benefit from the speed of dictation.
Form filling is an underrated application. Any time you need to enter text into structured fields — database records, customer relationship management systems, legal forms — dictation at the field level is faster than typing.
Steno is designed around these real-world use cases. It sits in your Mac menu bar and activates with a keyboard shortcut, placing transcribed text directly at your cursor in whatever application you are using. You can dictate into a Gmail compose window, a Notion document, a Slack message, or a spreadsheet cell without changing your workflow. For professionals who want to maximize their output across the full range of English writing tasks, it is one of the most practical tools available.
Starting Your Dictation Practice
The biggest barrier to adopting speech to text is simply inertia. Typing is a deeply ingrained habit, and switching to a new input method requires conscious effort for the first week or two. The way to make the transition stick is to pick one specific task — morning email responses, meeting notes, or daily journal entries — and commit to dictating that task for two weeks straight.
After two weeks, you will have built enough fluency that dictation starts to feel natural, and you will begin to notice all the other situations where it would be faster. Most people who stick with dictation for a month end up using it for the majority of their writing, not because they forced themselves to, but because it is genuinely faster and often produces better first drafts.
The average person speaks three times faster than they type. Speech to text in English simply closes the gap between the speed of thought and the speed of writing.
Download Steno at stenofast.com and start dictating in any Mac app within 30 seconds. No configuration required — just hold the hotkey, speak, and release.