NavodPeiris/mindWave

Issues

• add schizoprenia, paranoia detection using LLama2

  #18 opened 8 months ago by NavodPeiris
  0

• Google transcription is only accepting 30s audio

  #17 opened a year ago by NavodPeiris
  1

• 20s audio - processing time = 4min

  #16 opened a year ago by NavodPeiris
  0

• English transcription - Whisper Medium (small, tiny models are innaccurate), slow processing,

  #15 opened a year ago by NavodPeiris
  0

• train scream detection, cry detection models

  #14 opened a year ago by NavodPeiris
  1

• audio has noise

  #12 opened a year ago by NavodPeiris
  1

• test with multiple devices

  #10 opened a year ago by NavodPeiris
  1

• numbers not spelled by TTS

  #13 opened a year ago by NavodPeiris
  1

• Run 10 test cases with different ways.

  #4 opened a year ago by NavodPeiris
  1

• speaker recognition not working

  #6 opened a year ago by NavodPeiris
  1

• Voice activity detection low when talking quietly

  #7 opened a year ago by NavodPeiris
  1

• Use Pyannote speaker diarization 3.0.0

  #8 opened a year ago by NavodPeiris
  1

• modify to operate when more than 2 people talk

  #9 opened a year ago by NavodPeiris
  1

• fail on fd 51, errno: 104, "Connection reset by peer". suddenly while ESP32 streaming it crashed.

  #11 opened a year ago by NavodPeiris
  1

• add document queries through LLM

  #5 opened a year ago by NavodPeiris
  1

• use LLama2 for detecting if answer is related to the question

  #3 opened a year ago by NavodPeiris
  1

• add printing emotion recognition

  #2 opened a year ago by NavodPeiris
  1

• add emotion recognition

  #1 opened a year ago by NavodPeiris
  1