When it comes to voice assistants, the services that pop up in our minds are from top tech giants like — Apple’s Siri, Amazon Alexa, Google Assistant, and Microsoft Cortana. And we know they are collecting data by listening to the recordings their voice assistants capture and used them for target advertising.
Kinda break our privacy — the tool which we think meant to make our lives easier is capturing our private conversations, and we are still using them in our devices for years.
Are they are our only options? — definitely NO.
Let’s introduce you to the world’s first open-source voice assistant — Mycroft.
Yes, it’s an open-source project, so you don’t have to worry about your data being collected for target advertising. Mycroft values your privacy.
Why should involve with the Mycroft project?
Mycroft primarily based in Python, and the project is not at a state where it’s ready to gain mass adoption until in a year or so. And if you are a developer, contribute to the project, put some effort, file some bugs. Check out the official documentation and project source code is available on GitHub.
An essential feature presently a voice assistant needed is privacy, and Mycroft puts privacy at the forefront. Its default speech-to-text (STT) backend is Google’s STT service; also, there is the option to choose from a wide variety of providers, like IBM’s Watson, Mozilla, Kaldi, Microsoft Azure, and more. Besides, Mycroft proxies all STT requests that go through their servers for specific providers (Google being one of them).
When it comes to security, in Mycroft as a developer, you can implement VLANs, routing rules, firewall rules, and whatever other security that is relevant for your environment. Like if it’s on your network, you control what and who has access to your voice assistant.
Technologies used in Mycroft:
Mycroft is the name of a suite of software and hardware tools that utilizes natural language processing and machine learning to provide an open-source voice assistant. Mycroft system is built with different compounds, and it is modular. Here are the important technologies used in Mycroft:
- Wake Word detection — A Wake Word is a phrase you use to tell Mycroft you’re about to issue a command. By default, this is “Hey Mycroft”, but you can configure your own Wake Word.
- Speech to Text (STT) — Speech to Text software is used to take spoken words and turn them into text phrases that can then be acted on.
- Intent parser — An intent parser is a software that identifies what the user’s intent is based on their speech. An intent parser usually takes the output of a Speech to Text engine as an input.
- Text to Speech (TTS) — Text to Speech software takes written text, such as text files on a computer, and uses a voice to speak the text. Text to Speech can have different voices, depending on the TTS engine used.
- Middleware — The Mycroft middleware has two components, Mycroft Core: written in Python, is the core software that provides the ‘glue’ between other modules. And Mycroft Home and Mycroft API: where data on Users and Devices is held.
- Mycroft Skills — Mycroft Skills are like ‘add-ons’ or ‘plugins’ that provide additional functionality.
- Devices and Enclosures — Mycroft is designed to run on many different platforms – called Devices, with different hardware, called Enclosures.
Why use Mycroft?
First of all, Mycroft is open source and tools like Amazon Alexa, Google Assistant, Microsoft Cortana or Apple’s Siri are black boxes — so, you can’t look inside them. Most importantly, Mycroft uses opt-in privacy. This means it will only record what you say to Mycroft with your explicit permission.
Mycroft runs on a wide range of software platforms and hardware, and also, it is lightweight, so it will run on low-powered hardware, such as the Raspberry Pi 3.
If you are really worried about big tech companies gathering your data and used it for their profit, then supporting Mycroft is one of the necessary steps you have to take to stand against them since Mycroft is going to be the perfect open-source alternative to these top voice assistants.
