Projets-2015-2016-Sonotone

From air
Jump to: navigation, search

Subject: Sonotone

Supervisors:

  • Jérôme Maisonasse

Students:

  • HATTINGUAIS Julian (RICM4)
  • LECORPS Germain (RICM4)
  • VOUTAT Manuel (RICM4)


Subject summary

Week 1 (January 18th - January 24th)

Objectives

Documentation on sonotone (learning process)
Study of software already used for hearing aid (learning process)
Learning of a new programmation language : python with NUMPY and SCYPI modules

Work done

Assembling the needed documentation
A few prototypes and little program on Fourier transform and signal edition

Problems faced

Difficulties to apply a filter to an audio signal in python

Week 2 (January 25th - January 31th)

Objectives

Prototype of an equalizer filter (3 weeks job)
Learning of new concepts in Signal Processing

Work done

More prototypes about filters and signals edition
Learning of the concepts we need for peacking filter and z-transform

Problems faced

Difficulties to apply our theoritical knowledge of a numerical processing

Week 3 (February 1st - February 7th)

Objectives

Prototype of an equalizer filter (3 weeks job)
Apply a gain on a specific frequency (by peaking filter)

Work done

prototype of a peaking filter which works and which is performant

Problems faced

Difficulties to apply the gain the correct frequency

Week 4 (February 8th - February 14th)

Objectives

Prototype of an equalizer filter (3 weeks job)
Graphic equalizer

Work done

Graphic equalizer done!

Graphic Equalizer

Problems faced

None

Week 5 (February 15th - February 21st)

Objectives

Prototype of a parametric equalizer

Work done

We replaced peaking filters with shelf filters to improve the quality of the signal processing

Problems faced

Applying mutliple filters on one signal leads to several bugs and is less effective than shelf filters

Week 6 (February 29th - March 6th)

Objectives

Filtering in real time / Debug of parametric equalizer

Work done

Capture of the sound from a microphone and digitizing that sound via a C module.

Problems faced

Convert data from the C module to python data we can use in the equalizer.

Week 7 (March 7th - March 13th)

Objectives

Filtering in real time

Work done

Convertion of the C data to python data so we can apply filters to the signal with our equalizer.

Problems faced

No problem faced at this stage.

Week 8 (March 14th - March 20th)

Objectives

Filtering in real time

Work done

Learning and use of pyaudio module

Problems faced

Signal interference due to the microphone and maybe the python module

Week 9 (March 21st - March 27th)

Objectives

Filtering in real time/GUI (Design, IHM…)

Work done

Modification and amelioration of the filters Application of the filters in real-time signal Parsing of an XML file for options

Problems faced

Again, the interference but amplificate by the filters

Week 10 (March 28th - End)

Objectives

GUI (Design, IHM…)

Work done

Apply the GUI to the new real-time software

Problems faced

Difficulties to make a GUI in a simple way