A digital voice milestone was announced at the recent acoustics and speech conference in Caliornia when the Free DV Project's David Rowe VK5DGR copresented a paper describing a neural network that replaces traditional signal processing with machine learning.
In a recent post on the FreeDV website, David called the development: [quote] "the first known real-world deployment of a neural codec – an important milestone that the Ham community can be proud of." [endquote] He and programmer Jean-Marc Valin presented the details to attendees at the IEEE Signal Processing Society conference where David said it was well-received.
Instead of using the fixed algorithms of traditional digital voice, the FreeDV Radio Encoder, known as RADE V1, employs fully adaptive machine learning, producing a higher-quality result, developed using open source software.
Writing as a guest contributor to the Amateur Radio Digital Communications website in October, David noted that the technology [quote] "provides unprecedented speech quality and robustness for VHF/UHF land mobile radio applications." [endquote] The FreeDV project has been doing t his work with grant support from ARDC.
