According to the National Center for Voice and Speech, the average American speaks English at a rate of 150 words per minute. By comparison, according to Pocketables, a smartphone typist using the “two-thumb” method may average around 30 words per minute.
Why settle for basic typing when you can dictate your text up to five times faster? Sure, you can speed things up a bit with a keyboard that supports gesture typing, but even then you won’t get anywhere close to the same speed. That’s why you should consider using speech-to-text.
Note that speech-to-text isn’t only for convenience. It’s also an effective way to make your device more accessible. And we aren’t just talking about recording your voice as audio — speech-to-text takes voice input and converts it into text on-the-fly. Use it in conjunction with text messaging apps, for example!
Turn On and Set Up Speech-to-Text
The following instructions are based on a non-Samsung device running Android 5.1 Lollipop. Other versions of Android may differ slightly from step to step, but the overall procedure should be the same.
To set up speech-to-text, go to Device Settings, scroll down to the Personal section, and tap on Language & input. Scroll down to the Speech section and tap Voice input. Here you can choose between two voice input services.
- Basic Google recognition — Nothing more than the core speech-to-text recognition engine provided by Google.
- Enhanced Google services — Additional features on top of Google’s core speech-to-text recognition engine, including always-on voice monitoring and voice control for third-party apps.
While writing this post, I tried using both to see if there were any practical speech-to-text differences. The enhanced engine failed to identify any word I said, and no amount of tweaking fixed it. The basic engine didn’t have this issue, though I had to speak directly into the microphone for it to pick up. Your mileage may vary depending on your device and Android version.
Once you’ve picked an engine, tap the gear icon beside it to tweak its settings.
First, tap Languages and scroll down until you see the one that you need. Pay attention to the parentheses next to each language as these indicate accents. If you speak UK English but accidentally select US English, Google won’t be able to identify half the words you say.
Next, tap Hands-free and enable (or disable) whether you want speech-to-text to be recognized on external microphones that are connected either by a cable or by Bluetooth. This can come in handy when you want to voice-type text messages while driving, for example.
Next, tap Offline speech recognition, switch to the All tab, and download the language/accent packs that you want Google to recognize when your device is offline. This can also speed up speech-to-text conversion even while online since your voice no longer needs to be sent to Google’s servers for conversion. But if you need more device space, feel free to uninstall them.
Lastly, consider turning on text-to-speech. Speech-to-text is only half of it. It’s one thing to “speak” your text messages while driving to increase safety, but you should also “listen” to your text messages so you never have to take your eyes off the road.
How to Use Speech-to-Text on Android
Now that speech-to-text is set up and ready to go, let’s try using it.
Open up any app that normally requires typing: memo apps, SMS apps, to-do list apps, etc. Tap any text input area to bring up the on-screen keyboard. In the screenshot below, I’m using Gboard with a Colemak layout. Yours may look different (and some keyboards don’t support voice input).
Tap the microphone icon to switch the keyboard into voice input mode. At the bottom, tap and hold the Hold to talk button to begin voice recording. Speak into your microphone, then let go when you’re done. The converted text will be “sent” as if you had typed and tapped Send. (I personally wish it would just type the text and allow me to revise it, but oh well.)
When you’re done, tap the keyboard icon to revert to normal input mode.
What’s nice about speech-to-text is that this feature can be used anywhere the keyboard is used. SMS messages? To-do lists? Note-taking apps? Web searches? Google Search, Google Now, or Google Docs? Yes, everywhere.
To minimize frustrations, heed these tips:
- Speak clearly. Google’s recognition engine is good, but it isn’t perfect. Don’t speed too quickly and don’t slur your words together.
- Reduce background noises. If you’re driving with the windows down, or if you’re at an outdoor dog park, the ambient sounds will interfere with the clarity of your words. Cupping your mouth around the microphone might help, but don’t count on it.
- Keep using it. As with most things, speech-to-text gets better with practice. Not only does Google learn to better understand you, but you’ll learn how to deal with certain quirks and nuances.
The only downside is that holding the button to talk can be problematic for long-duration input. Imagine trying to record a lecture this way. (Good luck with that.) So while speech-to-text is great for simple sentences, for anything more involved, you should record voice notes with a third-party app instead.
One notable exception is the Google Keep app. It has a built-in feature that can translate voice to text. If you want long-form dictation in a non-audio format, this is the way to go.
Expand Your Voice Control on Android
Once you get comfortable talking to your Android device, you may want to graduate from speech-to-text to full-blown voice commands. Why bother tapping through all kinds of menus when you can just tell your smartphone or tablet what you want it to do?
For example, Google Assistant provides several features that make hands-free driving more convenient, and you should start using these nifty voice commands. (Especially for map navigation, which Waze isn’t as good for.) Other third-party apps are, slowly but surely, adding voice control too.
If you become proficient with Google Assistant, you can even control your entire device by voice. If you don’t like Google Assistant for whatever reason, maybe you’d prefer one of these other smart voice assistants instead.
How do you feel about using speech-to-text and voice control? Is it the future of mobile interfaces or just a passing fad? Let us know in the comments!