Deep SDR
Jump to navigation
Jump to search
This is an idea page for designs for an SDR project involving deep learning.
Implementation approach[edit]
- Build sw/hw radio to collect data
- feed training data to deep net
- voice recognition: extract transcript
- use existing OSS for recognition?
- voice semantics: extract call signs for labels
- handle unlabeled voices
- voice finger printing: match voices to call signs
- attempt fingerprinting radios: match voices to radios
- voice recognition: extract transcript
- program responses to specific voices
- greetings
- mailboxes
- RDF
Data collection[edit]
- hardware: RTL-SDR on a pc or pi
- receive on one or more repeater outputs
- bandwidth, segregate into channels
- squelch detection
- CTCSS recognition
- FM detection, audio volume detection
- signal strength
- filter
- remove ctcss
- cut at squelch transitions
- split into manageable chunk sizes
- save data
- timestamp, frequency
- demodulated data
- I/Q data (later?)
- voice recognition transcript
- codec2 compression??
Training and analysis issues[edit]
- bias in data: male / female
- supervised labeling
- unsupervised labeling from transcripts / check correlation for errors
- noise in sample --> issues in transcription
- labeled and unlabled samples