Mimic 3 release
Mycroft launched Mimic 3 June 29th, 2022! A privacy-focused open-source neural text-to-speech (TTS) engine. In even better news, Mimic 3 will be the default TTS engine on all Mark II devices.
Mike Hansen, Mycroft Senior Software Developer, sought to close the privacy loop by developing a TTS engine that sounds great and works offline. For more information on Mimic 3 check out its nerdy details on the Mimic 3 product page.
Mimic 3 can:
- Speak more than two dozen languages, with over 100 English voices available
- Run completely offline on devices like the Raspberry Pi 4
- Control speaking rate and variability
- Use SSML (Speaking Synthesis Markup Language) to switch voices, add pauses, and pronounce words phonetically within a single document
- And more…
Mimic 3 is for anyone who wants:
- To integrate a state-of-the-art text-to-speech system into just about anything;
- A personal, offline text-to-speech system that sounds better than ever;
- A premium text-to-speech cloud service, but also cares about the privacy of their data;
- To contribute their language expertise (or voice) for everyone’s benefit; or
- To hack on open source code.
Hardware
Preparations for manufacturing the Mark II continue. We are looking good on most parts. The speakers we were previously concerned about have shipped to us, expected on our dock in 2 weeks. We failed our Preliminary FCC Radio Frequency test back in mid May, but have since found the root causes and have solutions.
It is not uncommon to fail in the first test. Another test will run this Friday to confirm our solutions. When all goes well, this will give us our final FCC and CE certifications.
Once we have passed this test, we will order the SJ201s and displays for pre-production and production run one. The SJ201 fabricator has ordered the parts in anticipation of our order. We have completed the design on most of the parts for the plastic housing. The main housing mold is actually in process of being constructed. Some of the other housing parts may be impacted by the FCC fix. The next step are final drawings to be completed and sent to Zero Tolerance, our manufacturer of the molds, and provider of the plastic housings. We are a little behind where we wanted to be at this time due to the FCC test failure, but expect to still meet our September target date for shipping.
Hardware testing suite
Returning faulty hardware is expensive and frustrating for everyone involved. So to make sure that every Mark II coming off the production line is fully operational, we have been developing a complete suite of hardware tests. These will run on every single Mark II before it is packaged up and shipped to your door.
The tests cover all aspects of the Mark II hardware:
- Microphones
- Audio output
- Touch Display
- LED’s
- Hardware buttons
- Wi-Fi module
- GPIO pins;
- Even the cooling fan.
And all of this will be completed in under 60 seconds per device so that we can keep production rolling, particularly as we scale up the manufacturing runs.
Mycroft is hiring! Join our talented, dedicated team in our quest to provide the world the only open source, privacy-focused smart speaker. See available positions at Mycroft Careers.
Duane joined Mycroft March 2021. He has extensive experience in program management, IT, manufacturing, and as an entrepreneur. He is an enthusiast for skiing, playing guitar, music, biking, innovation, reading, learning, and good comedy (alright, even bad humour!). He is a believer, and excited about Mycroft, our staff, our product, our vision, our community, and our customers! Let’s all make it happen – great product by good people for good people.