Good evening all!
Around 2000/2001, I did some experiments with CMU Sphinx during an internship. Any of you who are interested in speech recognition will know that Sphinx was a major milestone in speech recognition technology.
I've just noticed that PocketSphinx ("a lightweight speech recognition engine, specifically tuned for handheld and mobile devices, though it works equally well on the desktop", https://github.com/cmusphinx/pocketsphinx) has been around for a while.
Bela's been a huge inspiration to me already (although I'm still waiting on some of the parts I ordered to ship, it's seems to have gotten into a bit of a muddle).
The potential of Bela+PocketSphinx has reignited some of my interests in the signal processing foundations of speech recognition, and I'd like to explore this more. So I have two questions:
Has anybody experimented with PocketSphinx on the BBB or Bela? (Or my other favorite SBC series, the Orange Pi?)
Has any anybody done any experiments with speech recognition technology at all with Bela, for example, computing MFCCs?
Any pointers or advice would be well appreciated. Thanks!!
Kindest regards,
ullbeking a.k.a. Andrew Nesbit