Python Ghostbox

   _____ _               _   _               
  / ____| |             | | | |              
 | |  __| |__   ___  ___| |_| |__   _____  __
 | | |_ | '_ \ / _ \/ __| __| '_ \ / _ \ \/ /
 | |__| | | | | (_) \__ \ |_| |_) | (_) >  < 
  \_____|_| |_|\___/|___/\__|_.__/ \___/_/\_\
                                             

Scans through AM/FM radio stations using SDR and Python.

Requirements

1. RTL-SDR or equivalent
2. GQRX
3. Python 3.4+

STT / TTS

1. torch
2. torchaudio
3. sounddevice
4. numpy
5. omegaconf
6. sentencepiece
7. soundfile

Usage

1. pip3 install -r requirements.txt
2. Start GQRX and enable remote control
3. Start UDP for speech to text if needed (reconfigure the UDP host to 127.0.0.1 in GQRX)
4. Run the script
   • python3 ghostbox.py --fm -i 0.15 --bounce

usage: ghostbox.py [-h] [-v] [--ip IP] [-p PORT] [--fm] [--am] [-i INTERVAL] [-s SQUELCH] [--random] [--forward]
                   [--backward] [--bounce] [--stt] [-w] [--tts] [-r] [-l] [-pb] [-ps PLAYBACK_SPEED]

Ghostbox

options:
  -h, --help            show this help message and exit
  -v, --version         display the program version
  --ip IP               gqrx IP
  -p PORT, --port PORT  gqrx port
  --fm                  enable FM radio scanning
  --am                  enable AM radio scanning
  -i INTERVAL, --interval INTERVAL
                        scanning interval in seconds
  -s SQUELCH, --squelch SQUELCH
                        squelch
  --random              random scanning
  --forward             forward scanning
  --backward            backward scanning
  --bounce              bounce scanning
  --stt                 enable speech to text
  -w, --wordlist        use a wordlist after audio is processed
  --tts                 enable text to speech
  -r, --reverb          apply reverb effect to TTS
  -l, --long-words      hide short words from the output
  -pb, --playback       playback transcript audio
  -ps PLAYBACK_SPEED, --playback-speed PLAYBACK_SPEED
                        playback speed

Speech to Text

The Ghostbox is capable of speech recognition by processing the UDP audio stream from GQRX. GQRX can be muted and audio will still be processed.

python3 ghostbox.py --fm -i 0.15 --bounce --stt -w --tts

macOS

1. Install homebrew
2. Install miniconda
   • brew install miniconda
   • conda init "$(basename "${SHELL}")"
3. Restart your shell
4. Install PyTorch
   • conda install pytorch torchvision torchaudio -c pytorch
5. Install Python requirements
   • pip3 install -r requirements.txt

Build GQRX for M1

If you get an error about libxml2, just ignore it and copy the GQRX.app file to your Applications folder.

• https://gist.github.com/Forst/db68106136be3380086e3c38be094d99

GQRX

• Disable Hardware AGC
• Gain 0.0 dB

Wordlist

Top 5000 common English words.

• https://github.com/filiph/english_words/

TimeseriesCache

• https://gist.github.com/jamesp/1385745/4736616d10551a6abc9c7c58725ee5b2f8de9f4d

Errors

UDP Bind Error

It's possible a copy of the Python script is still running in the background.

kill -9 $(ps -A | grep python | awk '{print $1}')

License

MIT