Anker has unveiled a new wearable voice recorder called the Soundcore Work. The device uses artificial intelligence to transcribe, summarize, and mark key parts of conversations. What makes it stand out is its incredibly small size; it is only 0.91 inches wide, making it smaller in diameter than a coin.
The Soundcore Work is set to launch in September, with a broader global release planned for later this year. It will cost $99.99(~Php5,600). That price includes what Anker calls “basic services.” The company will also offer an optional monthly subscription for $15.99(~Php900) that unlocks additional features, though specifics on those features have not been shared yet.
Despite its tiny size, the device is a bit thicker than a small coin. It weighs just 10 grams and can be worn on a lanyard or attached to clothing with a magnetic clip. Anker claims the battery lasts for up to eight hours on a single charge. The recorder comes with its own proprietary charger that has a built-in battery, which can extend the total usage time to 32 hours.
See also: How to turn yourself or any object into an action figure with Google Gemini
The recorder is equipped with two microphones. It does not record constantly. Instead, you tap it once to start recording a conversation or meeting. A double tap during a recording marks a specific moment as important, so it can be easily found later in the transcript.
The Anker Soundcore Work uses advanced AI models to process audio. It currently uses OpenAI’s GPT-4.1 technology, and Anker says it will eventually upgrade to GPT-5.0. It can transcribe speech in over 100 languages.
All transcripts and audio recordings are accessible through the Soundcore mobile app, where they can be searched or exported. However, the process does involve the cloud. When a user asks the app to transcribe or summarize a recording, the encrypted audio files are sent to the AI service provider’s cloud for processing. Anker states that these files are deleted from the cloud after the task is complete.
Anker also clarified that audio files are also temporarily sent to the cloud during the transcription process, not only when they are being shared with another person. Files are also sent to the cloud when a user chooses to share a recording.