Speech to Text
See conversations as text, instantly. For noisy places, important moments, or anytime you'd rather read along.
By Lilly Seay · Updated July 2026
Want 100% private captions?
The Hearing Buddy app processes everything on your device — your conversations never leave your phone. Plus speaker identification, conversation summaries, and Apple Watch support.
When Speech to Text Helps
Crowded Places
Restaurants, airports, coffee shops — background noise makes it hard for anyone to hear. Speech to text cuts through the chaos so you catch every word of the conversation, even when the room is loud. About 37.5 million American adults report some trouble hearing, according to the NIDCD, and noisy environments are the #1 challenge they face. If noise is a daily struggle, our hearing aid comparison can help you find one built for it.
Important Conversations
Doctor's appointments, work meetings, lectures — some conversations are too important to miss. Real-time transcription gives you a visual backup, so you can focus on understanding instead of straining to hear. A research review synthesizing more than 100 studies found captions improve comprehension for deaf and hard of hearing viewers and hearing people alike (Gernsbacher) — captions help everyone.
Listening Is Exhausting
Straining to follow speech all day is genuinely tiring — research in Ear and Hearing found adults seeking help for hearing difficulty report significantly more fatigue than population norms, tracking how hard listening feels more than what the audiogram says. Reading along takes real load off. If social evenings wipe you out, captions aren't a crutch — they're energy management.
The Right Captions for Every Situation
One tool doesn't fit every moment. Here's the honest map for mid-2026 — what to reach for, where.
In-person conversations
On a computer, this page works instantly. On your phone, the Hearing Buddy app (iPhone) captions with speaker ID entirely on-device, and Android’s Live Transcribe is the purpose-built free option. Put the phone close to whoever is talking — distance is the enemy.
Phone calls
The most underused benefit in hearing care: federally funded captioned-call services are completely free for people with hearing loss. Apple Live Captions and Android Live Caption also caption calls on the phone itself. Full details in the phone calls section below.
Video meetings
Zoom, Microsoft Teams, and Google Meet all have built-in captions any participant can turn on for themselves — in Teams it’s under "Language and speech," in Meet it’s the CC button, and in Zoom the host may need to enable automated captions once in settings.
TV, videos & media
Android’s Live Caption and Windows 11’s Live Captions transcribe whatever your device plays, on-device. This fall, iOS 27 adds Generated Subtitles that automatically caption any video that shipped without them. For the living-room TV itself, see our TV hearing solutions guide.
Class, work events & lectures
For anything high-stakes, ask for CART — a human real-time captioner. It’s a recognized accommodation employers and colleges are legally required to consider (details in the rights section below). For casual lectures, Otter or Google Recorder save a searchable transcript.
Another language
Live captioning and live translation are different features. Teams and Meet offer translated captions on paid plans, Samsung’s Galaxy AI does two-way in-person translation on recent Galaxy phones, and iOS 27 will translate existing video captions this fall.
Free Live Caption Tools Compared (July 2026)
Every option here is reputable and genuinely free at the tier shown — the differences are platform, privacy, and what happens to your transcript. Details verified against each vendor's official documentation.
| Tool | Platform | Best for | Cost | Keep in mind |
|---|---|---|---|---|
| This page (Hearing Buddy web) | Any browser | Instant captions on any computer, nothing to install | Free, no sign-up | One speaker at a time; needs internet in most browsers; nothing is stored by the page. |
| Hearing Buddy app | iPhone + Apple Watch | In-person conversations with color-coded speaker ID | Free | Everything processes on-device — audio never leaves your phone. Name and question alerts tap your wrist on Apple Watch. |
| Apple Live Captions | iPhone 11+, iPad, Mac | System-wide captions: calls, FaceTime, media, nearby speech | Free, built in | On-device; language support expanded well beyond English in the iOS 26 cycle. Captions are ephemeral — no saved transcript. |
| Google Live Transcribe | Android 8+ | Face-to-face conversations on Android | Free | 70+ languages; offline language packs on Android 12+. Transcript history auto-deletes after 3 days. |
| Google Live Caption | Pixel 2+ and select Android | Captioning calls, videos, and podcasts playing on your phone | Free | Fully on-device. Captions device audio only — for room conversation, use Live Transcribe instead. |
| Windows 11 Live Captions | Windows 11 (22H2+) | Captioning anything on a PC, including your own mic audio | Free, built in | Press Win+Ctrl+L. On-device; enable "Include microphone audio" to caption in-room speech. Display-only, no transcript. |
| Otter.ai | Web, iOS, Android | Meetings and classes where you want a saved, searchable transcript | Free: 300 min/month, 30 min per conversation | Account required and cloud-based — audio is uploaded to Otter’s servers. The free cap runs out fast as a daily conversation aid. |
| Ava | iOS, Android, desktop, web | Group conversations — everyone joins from their own phone, each speaker captioned by name | Free basic plan (40-minute sessions) | Built by and for the deaf/HOH community. Cloud-based, account required; a human "Scribe" tier offers ~99% accuracy with advance booking. |
| Google Recorder | Pixel phones | Lectures and long meetings you want recorded and searchable | Free (preinstalled) | Live transcription is on-device on Pixel 6+; automatic speaker labels. Pixel-only. |
| Zoom / Teams / Meet captions | Inside each meeting app | Video meetings without installing anything extra | Free for participants | Turn on per-user (Teams and Meet; Zoom hosts may need to enable automated captions once). Translated captions generally require paid plans. Cloud-based. |
Searching for "Live Transcribe for iPhone"? Google's app is Android-only — on iPhone, the equivalents are Apple's built-in Live Captions, this web page, or the Hearing Buddy app. And coming this fall: iOS 27 adds Generated Subtitles, which automatically captions any video that shipped without them.
Captioned Phone Calls Are Free — Really
Phone calls are the hardest listening situation there is: no faces, no lip movement, compressed audio. Here's the part almost nobody knows — the US has a federally funded program that captions your calls for free.
How the Free Program Works
Under the ADA, the FCC ensures captioned telephone service (IP CTS) is available in all 50 states and territories, paid for by the federal TRS Fund — there is no cost to you (FCC). To qualify, you self-certify that hearing loss makes the phone hard for you, and register with a provider (name, address, date of birth, last four of your SSN). Then your calls simply appear with captions — on a captioned phone, your smartphone, or a computer.
Who Provides It
As of July 2026, nine companies hold FCC certification, including InnoCaption (choose live stenographers or automated captions, and switch mid-call), CaptionCall by Sorenson (free captioning phone plus a mobile app), Hamilton CapTel, ClearCaptions, and Rylo (formerly Nagish — fully automated, with transcripts kept on your device). The full certified list lives on fcc.gov.
Built Into Your Phone
No sign-up required for the basics: Apple Live Captions captions phone calls and FaceTime on iPhone (with speaker attribution in FaceTime), and Android's Live Caption captions calls on Pixel and select phones — both on-device. There's also RTT (real-time text), built into modern smartphones: text flows character-by-character during a call with no send button, works alongside voice, and reaches 911. And where local 911 centers support it, you can text 911 directly — if not, you'll get an automatic bounce-back telling you to reach help another way.
How Accurate Is Live Speech to Text?
Vendors love to claim "99% accuracy" with no caveats. Here's what the research actually shows — and how to get the best from any caption tool.
The Honest Numbers
On clear, clean speech, leading recognition systems get roughly 97 of 100 words right and nearly match professional human transcribers (Whisper evaluation). But on real dinner-party conversation recorded from across the table — overlapping voices, distance, clatter — the best research systems still missed about one word in three (CHiME-6). Both numbers are true. The environment is the difference.
What Drags Accuracy Down
Background noise, distance from the microphone, people talking over each other — Google's own documentation warns overlapping speakers "may be interpreted as background noise and ignored." Specialized vocabulary hurts too: a study of clinical dictation found 7.4 errors per 100 words in raw output, including medication names. Fast talkers and thick crosstalk finish the list.
Not Equal for Every Voice
A 2020 study in PNAS tested five major commercial recognizers and found nearly double the error rate for Black speakers versus white speakers (35% vs 19% average WER), driven by underrepresentation of dialect phonology in training data (Koenecke et al.). If captions serve your voice or your family's voices worse, that's a real, documented limitation of the technology — not you.
Getting Your Best Results
Put the microphone close to whoever is speaking — it's the single highest-impact fix. One speaker at a time beats crosstalk every time. In loud rooms, a directional or remote microphone captures the voice at its source, the same principle assistive listening systems use. And read the trend, not each word: meaning usually survives even when individual words glitch.
Where Does Your Audio Actually Go?
"Private" means different things to different tools. A plain-language map — including the honest answer about this page.
Truly On-Device
Apple Live Captions, Android's Live Caption, Windows 11 Live Captions, the Hearing Buddy app, and Pixel's Recorder (Pixel 6+) all process audio entirely on your device — nothing is sent to a server. These are the right choice for medical, legal, and any conversation you'd rather keep in the room.
Cloud-Processed
Otter, Ava, and the built-in captions in Zoom, Teams, and Meet send audio to servers for recognition. This browser tool sits in between: the page itself never stores or uploads anything, but most browsers' speech recognition sends audio to the browser's recognition service — that's how the Web Speech API works today, per MDN (fully on-device browser recognition is emerging but still experimental).
What Sticks Around
Retention varies wildly: this page keeps nothing once you leave, Apple and Windows captions vanish as they scroll, Android's Live Transcribe auto-deletes its on-device history after 3 days, while Otter and Recorder are built to keep transcripts. Google explicitly notes Live Transcribe isn't a HIPAA-covered product — for sensitive appointments, use an on-device tool or ask for human CART captioning, which is your right under the ADA.
Captions Are Your Right, Not a Favor
Apps are wonderful, but sometimes the answer is a human captioner paid for by someone else — because the law says so.
At Work
Employers with 15+ employees must provide reasonable accommodations under the ADA, and the EEOC's guidance explicitly lists CART (real-time captioning) and captions on virtual meeting platforms among the examples (EEOC). Put the request in writing to HR.
At Businesses & Healthcare
Businesses serving the public must communicate as effectively with you as with anyone else, and the regulations explicitly list "real-time computer-aided transcription services" as an auxiliary aid — and prohibit charging you for it (ADA.gov).
At College
Schools receiving federal funds must ensure no student misses out for lack of auxiliary aids like captioning under Section 504 — and a student "may not be required to pay part or all of the costs" (Dept. of Education). The disability services office is your starting point.
At the Movies & on TV
Movie theaters showing digital films with captions must provide caption viewing devices — a minimum of 4 to 12 devices depending on theater size, per DOJ rule. On TV, the FCC requires captions to be accurate, synchronous, and complete, and takes written complaints within 60 days of a problem.
How Speech to Text Technology Works
Browser-Based Recognition
This tool uses the Web Speech API built into modern browsers like Chrome and Safari. When you grant microphone access, your browser's speech recognition engine converts audio to text in real-time. No additional software, plugins, or downloads required — just open the page and start talking.
Accuracy and Limitations
Browser speech recognition handles clear speech in quiet environments very well. Accuracy decreases with heavy accents, multiple overlapping speakers, or excessive background noise. For the most accurate experience, use a quality microphone and minimize ambient noise. The Hearing Buddy app offers enhanced accuracy with on-device processing.
Speech to Text vs. Live Captions
This browser tool transcribes one speaker at a time. If you need multi-speaker captions with speaker identification — where each person's words are labeled and color-coded — the Hearing Buddy app provides that with on-device processing, meaning your conversations stay completely private.
How This Compares to Built-In Captions
Captions on Your Phone
Android's Live Transcribe and Apple's Live Captions are excellent system-wide options that caption what your phone hears or plays. They're worth turning on if you own the device. This web tool fills a different gap: it works instantly on any computer — a work laptop, a library machine, a friend's browser — with nothing to install or set up.
What's New from Apple
Two things worth knowing: Name Recognition — which notifies you when your name is called — already shipped in the iOS 26 cycle under Settings → Accessibility → Sound Recognition. And this fall, iOS 27 adds Generated Subtitles, which automatically creates synced captions for any video that shipped without them and can translate existing captions into other languages, per Apple's iOS 27 preview. For TV dialogue in particular, see our TV hearing solutions guide.
Where This Tool Fits
Built-in captions shine for media playback; this page shines for live, in-person moments when you just need words on a screen right now. And captions pair well with the right device — if your hearing loss is more significant, take a look at the best hearing aids for severe hearing loss to see what pairs well with captions.
Questions & Answers
This tool uses your browser’s built-in speech recognition to convert spoken words to text in real-time. Just tap the button, allow microphone access, and start talking — your words appear on screen instantly.
Yes, completely free. No account needed, no credit card, no hidden fees. We built this tool to help people who need speech to text access it easily.
Here’s the honest answer: this page never stores or uploads your transcript, but browser speech recognition itself typically sends audio to your browser’s recognition service for processing — that’s how the Web Speech API works in most browsers today (MDN). If you need captions where audio truly never leaves your device, use an on-device tool: the Hearing Buddy app, Apple Live Captions, Android Live Caption, or Windows 11 Live Captions.
It works on most modern mobile browsers including Chrome and Safari. For the best mobile experience with additional features, we recommend the Hearing Buddy app.
Anyone who wants to see speech as text. Maybe you’re in a loud coffee shop, sitting in on a lecture, or just process information better by reading. It’s also great for people who are hard of hearing or deaf — live captions help everyone catch every word.
This web tool currently listens for English (US). If you need other languages, Android’s Live Transcribe supports 70+ languages and dialects, Apple Live Captions added languages including Spanish, French, German, Japanese, Korean, Mandarin, and Cantonese in the iOS 26 cycle, and Windows 11 Live Captions supports 20+ after downloading language files.
The transcript stays on screen while you’re on the page, and you can select and copy the text anytime before you leave — nothing is stored or uploaded by this tool. If you’d like conversation summaries you can look back on later, the Hearing Buddy app creates those privately on your device. For meetings, Otter and Google Recorder are built around saved transcripts.
No — Live Transcribe is Android-only, and that surprises a lot of iPhone owners looking for it. The good news: iPhone users have real equivalents. Apple’s built-in Live Captions (Settings → Accessibility → Live Captions, iPhone 11 and later) captions calls, FaceTime, media, and nearby conversation on-device; this web page works in Safari; and the Hearing Buddy app adds speaker identification and Apple Watch alerts with fully on-device processing.
Two ways. First, built-in: Apple Live Captions captions phone calls and FaceTime on iPhone, and Android’s Live Caption captions calls on Pixel and select phones. Second — and better for heavy phone users — the FCC’s captioned telephone service (IP CTS) is completely free for people whose hearing loss makes phone calls hard: federally funded, with apps and captioned phones from providers like InnoCaption, CaptionCall, Hamilton CapTel, ClearCaptions, and Rylo. You just self-certify your hearing loss when you sign up. See the phone calls section above for the full picture.
On clear speech in a quiet room, the best systems now get roughly 97 of 100 words right — nearly matching professional human transcribers. Real life is harder: research systems tested on real dinner-party conversation with distant microphones still missed about one word in three (CHiME-6). Noise, distance, crosstalk, accents, and jargon all pull accuracy down. Treat live captions as a powerful assist, not a court transcript — and see our accuracy section above for how to get the best results.
Overlapping speech is the hardest thing you can ask of any speech recognizer — Google’s own documentation warns that multiple people talking at the same time may be interpreted as background noise and ignored. The practical fixes: ask people to speak one at a time, put the phone near whoever has the floor, or use a tool built for groups — Ava links everyone’s phones into one captioned session, and the Hearing Buddy app labels speakers with color-coded speaker ID.
This web tool needs an internet connection in most browsers. But several on-device tools work offline: Windows 11 Live Captions and Apple Live Captions process locally, Android’s Live Transcribe works offline for languages you’ve downloaded (Android 12+), and the Hearing Buddy app runs entirely on-device. Handy on planes, in basements, and anywhere coverage dies.
They work, with caveats — accuracy drops as noise rises, for every tool. Stack the deck: sit against a wall away from the kitchen, put the phone close to the person you most need to hear, ask your group to take turns, and consider a directional or remote microphone — the same trick assistive listening systems use, capturing the voice at its source. The trend of the conversation usually stays readable even when individual words glitch.
Yes — and you have backup rights. Under the ADA, healthcare providers must communicate as effectively with you as with hearing patients, and real-time captioning is an explicitly listed auxiliary aid they may need to provide at no charge to you (ADA.gov). One privacy note: cloud-based caption apps upload audio to servers, which matters for medical conversations — prefer an on-device tool, or ask the provider to arrange CART.
Yes. Employers with 15 or more employees must provide reasonable accommodations under the ADA, and the EEOC’s guidance explicitly lists CART and captioning on virtual meeting platforms as examples. Colleges receiving federal funds must provide auxiliary aids like captioning under Section 504 — and may not require you to pay for them. Start with HR or the disability services office, in writing.
Very much so — they solve different problems. Hearing aids make speech louder and clearer, but when word discrimination is the issue, amplification alone can’t fill in consonants you can’t resolve. Captions give your eyes the words your ears miss, which is why many hearing aid wearers use both. Research on captions consistently shows comprehension benefits across the board, including for people with hearing devices.
Increasingly, yes — but usually behind a paid tier. Microsoft Teams offers translated captions with Teams Premium, Google Meet offers them on paid Workspace plans (70+ languages), Samsung’s Galaxy AI does two-way in-person translation on recent Galaxy phones, and iOS 27 will translate existing video captions when it ships this fall. For same-language captioning, the free built-in tools above cover most needs.
Never miss a word again
Whether you're in a noisy coffee shop, an important meeting, or just prefer reading along — we've got you covered.
Available on iPhone, Apple Watch, and Mac