Nothing launches Essential Voice, an AI dictation tool for smartphones supporting more than 100 languages

Nothing is expanding its AI lineup not with yet another chatbot but with a practical feature that could come in handy every day. The company has introduced Essential Voice, a built-in dictation tool for smartphones that turns natural speech into neatly formatted text. The bet is not just on recognizing words but on making mobile input more convenient: the service removes filler words, works in any app and prepares text in a form ready to send, save as a note or use as a work message.

Nothing announced the launch on April 24, 2026. Essential Voice is built in at the system level and, according to the company, should spare users from switching between separate dictation apps and the phone's main interface. The tool takes voice input, converts it to text and immediately cleans up the result. This goes beyond basic transcription: the system removes habitual pauses and interjections such as "uh" and "um" so the output is a cleaner message.

For cases where the same data is entered over and over, there are custom voice shortcuts. For example, you can bind a short command to a full address, a link, a boilerplate phrase or a recurring snippet of text.

The feature is currently available on the Nothing Phone (3). For the Phone (4a) Pro, release is planned by the end of April 2026, and for the Phone (4a), in May.

Dictation can be launched in two ways: with a dedicated Essential key on compatible devices or directly from the keyboard. That detail matters, because Nothing is trying to build voice input into the ordinary user flow rather than leave it as a niche feature buried deep in settings.

The company also stresses speed: by its estimate, the average user types about 36 words per minute on a smartphone, but can speak roughly four times faster. That claim underpins Essential Voice's positioning as a productivity tool rather than a decorative AI add-on.

Another notable capability is translating text from one language to another as you dictate. At launch, Essential Voice supports more than 100 languages, which makes it potentially useful not only for personal messaging but also for international work, travel and communicating with clients.

In upcoming updates, Nothing also promises to add editing-style settings by app category. The idea is that the AI could process the same voice draft differently depending on context: more formally for work tasks and more casually for messages. If this part is implemented well, dictation will start competing not only with ordinary speech recognition services but also with products that combine transcription, editing and reply templates.

Nothing is not alone in the market. AI dictation has become a fast-growing segment in recent years: Wispr Flow, Superwhisper, Willow, Monologue and other tools have appeared alongside earlier products, and new ones launch almost every week. Google recently introduced its own dictation solution as well, so interest in the category clearly extends beyond startups.

But Nothing has one important difference: it is among the first to try offering such a feature not as a standalone app but as a system layer inside the smartphone. That integration may prove its main advantage, because users do not need to change how they work. Voice simply becomes another standard input method next to the keyboard.

For Nothing, this is a small but telling announcement. The company does not promise to reinvent mobile AI; it takes a specific everyday scenario and makes it faster. If Essential Voice really works reliably in any app, cleans up text well and switches between languages smoothly, voice input has a chance to become a mass habit rather than a rarely used feature "in case your hands are busy."