Earlier quoted context omitted.
What company is this if you dont mind saying?
I assume he is talking about Planet Labs ( https://www.planet.com/ ) - a very cool service.
Do you happen to know what the pricing looks like?
Listening for illegal logging chainsaws using TensorFlow
91–100 of 128 posts
Re: Listening for illegal logging chainsaws using TensorFlow
#92Logging Man, I must have computers on the brain. This isn't about noisy server logs at all.
...but now that you point it out, I would totally try AI log analytics.
Re: Listening for illegal logging chainsaws using TensorFlow
#93They're streaming audio 24x7 all over the Amazon, but I can't get a signal in the office hallway.
Well, Amazon is already peered with major ISPs;)
Re: Listening for illegal logging chainsaws using TensorFlow
#94If a tree falls in the forest, TensorFlow will hear it.
> Hot deals on chainsaw chains in your area, click here for 50% discount
Re: Listening for illegal logging chainsaws using TensorFlow
#95Re: Listening for illegal logging chainsaws using TensorFlow
#96Does ShotSpotter have patents or something? It seems off that they aren't mentioned, but they're definitely prior art for this. Hopefully it works better for logging, which is to mean: at all.
It works pretty well, as demonstrated here in an event within the last couple of weeks in Peru:
https://twitter.com/lilitjejauregui/status/97585519352188518...
Re: Listening for illegal logging chainsaws using TensorFlow
#975 years from now: "listening for illegal speech in city streets using TensorFlow"
do you really think that happens?
no way TensorFlow is going to be alive in 5 years!
Re: Listening for illegal logging chainsaws using TensorFlow
#98TechCrunch just posted a slightly broader article on the project:
https://techcrunch.com/2018/03/23/rainforest-connection-enli...
Re: Listening for illegal logging chainsaws using TensorFlow
#99Where do I begin if I want to set something like this up? As an experiment - my neighborhood in India has a problem with loud motorbikes that are illegally modified. Hypothetically if I wanted to use this approach (there are probably simpler ones in this specific case), where do I start? Do I just start recording gigabytes of audio data every day and start doing some kind of FFT on the data and manually tagging cases…
I'd start with the end goal. Is the acoustic signal from the motorbikes likely to be traceable to particular bikes? If so, would the local police actually be interested in following up?
There is value in quantifying just how noisy the bikers are and for just how long they disturb the neighbourhood, but you'll probably need the perpetrators to be people the police are able and willing to stop to get anywhere.
Re: Listening for illegal logging chainsaws using TensorFlow
#100> We’ve hidden modified smartphones powered with solar panels—called “Guardian” devices
Question, how is the electrical PV power budget vs load budget calculated for these? Having implemented PV projects ranging in size from 300W to 30kW, I'm very interested to know what their predicted Wh per day generated from the solar cell is vs. what the load will consume (what's the actual load of the cellphone, how often does it wake up every day, etc).