Using AI To Pull Call Signs From SDR-Processed Signals

AI is currently popular, so [Chirs Lam] figured he’d stimulate some interest in amateur radio by using it to pull call signs from radio signals processed using SDR. As you’ll see, the AI did just okay so [Chris] augmented it with an algorithm invented for gene sequencing.

Radio transmitting, receiving, and SDR hardwareHis experiment was simple enough. He picked up a Baofeng handheld radio transceiver to transmit messages containing a call sign and some speech. He then used a 0.5 meter antenna to receive it and a little connecting hardware and a NooElec SDR dongle to get it into his laptop. There he used SDRSharp to process the messages and output a WAV file. He then passed that on to the AI, Google’s Cloud Speech-to-Text service, to convert it to text.

Despite speaking his words one at a time and making an effort to pronounce them clearly, the result wasn’t great. In his example, only the first two words of the call sign and actual message were correct. Perhaps if the AI had been trained on actual off-air conversations with background noise, it would have been done better. It’s not quite the same issue, but we’re reminded of those MIT researchers who fooled Google’s Inception image recognizer into thinking that a turtle was a gun.

Rather than train his own AI, [Chris’s] clever solution was to turn to the Smith-Waterman algorithm. This is the same algorithm used for finding similar nucleic acid sequences when analyzing genes. It allowed him to use a list of correct call signs to find the best match for what the AI did come up with. As you can see in the video below, it got the call signs right.

Continue reading “Using AI To Pull Call Signs From SDR-Processed Signals”

Listen to the Netherworld with Artificial Intelligence

It’s that time of year again, and with Halloween arguably being the hacker’s perfect holiday, we’re starting to see a tick up in projects with a spooky theme. Most seem to do with making some otherwise tame Halloween decorations scarily awesome, but this is different — using artificial intelligence to search for ghosts.

It seems like [Matt Reed]’s “DeepWhisper” project is meant to be taken as light-hearted fun for the spooky season, but there may be a touch of seriousness to his efforts to listen in on ghostly conversations. The principle behind this is electronic voice phenomena (EVP), whereby the metabolically and/or dimensionally challenged are purported to influence electronic systems, resulting in heavily processed audio clips that seem to have a whispered endearment from the departed or a threat from a malevolent spirit. DeepWhisper takes this a step further by using a Raspberry Pi to feed audio into the Google Cloud Speech API for analysis. If anything is whispered in one of the 110 or so languages Google knows, it’ll get displayed on a screen. [Matt] plans to set DeepWhisper up in the aptly-named Butchertown section of Nashville and live-stream the results next week.

It’ll be interesting to see what Google’s neural network makes out of the random noise it will probably only ever hear. And [Matt] is planning on releasing his code for all to see, so there may be some valuable cloud techniques to learn from DeepWhisper. But in the unlikely event that he does discover ghosts, it’s nice to know you can have the tools and the talent to bust ’em.

Continue reading “Listen to the Netherworld with Artificial Intelligence”

Seeed Studio’s ReSpeaker Speaks All the Voice Recognition Languages

Seeed Studio recently launched its third Kickstarter campaign: ReSpeaker, an open hardware voice interface. After their previous Kickstarted IoT hardware, such as the RePhone, mostly focused on connectivity, the electronics manufacturer from Shenzhen now tackles another highly contested area of IoT: Voice recognition.

The ReSpeaker Core is a capable development board based on Mediatek’s MT7688 WiFi module and runs OpenWrt. Onboard is a WM8960 stereo audio codec with integrated 1W speaker/headphone driver, a microphone, an ATMega32U4 coprocessor, 12 addressable RGB LEDs and 8 touch sensors. There are also two expansion headers with GPIOs, I2S, I2C, analog audio and USB 2.0 and an onboard microSD card slot.

The latter is especially useful to feed the ReSpeaker’s integrated speech recognition engine PocketSphinx with a vocabulary and audio file library, enabling it to respond to keywords and commands even when it’s not hooked up to the internet. Once it’s online, ReSpeaker also supports most of the available cloud based cognitive speech recognition services, such as Microsoft Cognitive Service, Amazon Alexa Voice Service, Google Speech API, and Houndify. It also comes with an SDK and Python API, supports JavaScript, Lua and C/C++, and it looks like the coprocessor features an Arduino-compatible bootloader.

The expansion header accepts shield-like hardware add-ons. Some of them are also available through the campaign. The most important one is the circular, far-field microphone array. Based on 7 XVSM-2000 respeaker_meow2digital microphones, the extension board enhances the device’s hearing with sound localization, beam forming, reverb and noise suppression. A Grove extension board connects the ReSpeaker to the Seeed’s current lineup on ready-to-use sensors, actuators and other peripherals.

Seeed also cooperates with the Meow King Audio Electronic Company to develop a nice tower-shaped enclosure with built-in speaker, 5W amplifier and battery. As a portable speaker, the Meow King Drive Unit (shown on the right) certainly doesn’t knock your socks off, but it practically turns the ReSpeaker into an open source version of the Amazon Echo — including the ability to run offline instead of piping everything you say to Big Brother.

According to Seeed, the freshly baked hardware will ship to backers in November 2016, and they do have a track-record of on-schedule shipped Kickstarter rewards. At the time of writing, some of the Crazy Early Birds are still available for $39. Enjoy the campaign video below and let us know what you think of think hardware in the comments!