How is a smart voice recorder ‘smart’, exactly? It is a perfectly natural question; after all, plenty of things are claimed to be ‘smart’ these days: cars, TVs, houses, watches, light bulbs, et cetera. So when you see a recorder labelled ‘smart’, the picture in your head is vague at best.
In this article, we are going to look at Senstone, a smart wearable recorder with an unusual design. The questions we are going to answer are: what are the features that make it stand out among the crowd? How is it actually smart? What is the vision behind the features Senstone has and what do they bring to the table?
Without a doubt, the most noticeable feature of Senstone is that, unlike most recording devices, it processes your recordings and organises them into a system. It does store the audio files, too, but they are just the tip of the iceberg, and the most interesting things happen in the app itself. In a way, Senstone makes ‘smart’ in its description refreshingly literal.
At its core Senstone is built around the system of natural language processing (NLP) – complex structures of algorithms specifically programmed to process and make sense of human language. The most well-known examples of NLP at work would be AI assistants that accept voice input like Siri and Alexa. Natural language processing is the very thing that lets Senstone build the deeper understanding of its user’s linguistic habits and transcribe the recordings.
For every voice entry, Senstone automatically creates a transcript that can be viewed and modified via the app. The accuracy of speech-to-text conversion holds up to the scrutiny and stands at 95% or higher – it is just as accurate as Siri.
As the transcript is being created, artificial intelligence inspects the text for accidental mistakes and irregularities, acting as an autocorrect. The AI also breaks the text into sentences and paragraphs, taking care of the syntax. Ideally, you would not have to edit anything at all unless you wanted to add a remark or expand on a topic. Editing can be done in the app.
As a part of its NLP, Senstone uses advanced semantic speech processing technology which takes into account the minute details of how users communicate their thoughts: the most frequently used words, sentence structure, favourite expressions and phrases. Capable of learning and self-improvement, the AI matches every incoming sentence and its parts to the patterns it has already learned and makes corrections where necessary.
This is the way Senstone learns – from experience. As you keep using Senstone, the artificial intelligence learns and improves further and further. Basically, what you get can be called a personal smart voice assistant.
Another feature worth mentioning is a personal dictionary. Again, it can be accessed from the Senstone app. The dictionary lets you add new words, such as particularly rare proper nouns or technical terms, to the list for AI to process and recognise. This ensures that Senstone will be translating the words into text properly and not attempt to correct them into something statistically more common.
The personal dictionary will be especially useful for health care workers, university students, writers, and other professionals who tend to make heavy use of specialised terms. It is intended as yet another option to create personalised experience and make interactions with Senstone as streamlined as possible.
One more interesting feature is entry tagging.
Senstone app allows for #conventional #tags as well as geotags, also known as location tagging. The tags can be added and edited at any time, and multiple tags can be applied to a single entry. The option is rather flexible, and this means you can set up a custom labelling system where some tags overlap, essentially sorting recordings by topic and/or location.
Regarding software, the final major advantage of Senstone is connectivity. It connects to the cloud to make back-ups, it syncs with your phone’s OS, and both audio recordings and transcripts are easy to share via social networks or basic copy & paste. Minimalistic and robust infrastructure makes it simple to distribute recordings quickly and efficiently.
In general, built-in tagging, highlights, time stamps, sharing options, and robust search functionality come together to create a system that can be tweaked and modified to anyone’s personal liking. Senstone certainly has a lot to offer, and personalisation is one of its strengths.
When it comes to hardware, Senstone takes on the ergonomic approach. The recorder is a wearable device the size of a button, shaped to resemble a fashionable accessory rather than a gadget. Since Senstone is designed as a personal voice recorder, it is recommended that users clip it to their collar or wear it as a pendant around their neck. A miniature LED light prevents stealth recording by giving a flash every time you start using the recorder.
To start a new recording, the user taps the device. There is only one button. Apart from it, Senstone has no other buttons, wires, or screens. Once the recording starts, the device responds to voice commands; other than that, settings, entries, and other features are accessible from the app.
Audio recordings in the MP3 format are stored in the cloud and on the user’s phone. Up to 2.5 hours of audio can be stored without access to the phone, and Senstone has no issues working offline.
The reason for Senstone completely hands-free is the productivity-based philosophy behind its creation.
Advocating focus and efficiency, Senstone refuses to accept keyboard as the default input method and treats manual recording – all of it, including touch screens – as a thing of the past which has been made obsolete by the current advances in speech recognition. Senstone defines itself as a tool to eliminate distractions and improve the productivity of its user, but the bigger picture of its vision is the future where voice and not text is the skeleton key to human–computer interactions.
All of the above contributes to the image of Senstone as a smart voice recorder, and it fits the definition of ‘smart’ tech. Following the trend for wearable devices, it offers portability, as well as accurate voice processing and rich functionality.
