Hearing and Deafness 2 Ear as a frequency

  • Slides: 30
Download presentation
Hearing and Deafness 2. Ear as a frequency analyzer Chris Darwin

Hearing and Deafness 2. Ear as a frequency analyzer Chris Darwin

Frequency: 100 -Hz Sine Wave 1. 0 Spectrum 0 Amplitude against frequency -1. 0

Frequency: 100 -Hz Sine Wave 1. 0 Spectrum 0 Amplitude against frequency -1. 0 0 Time (s) Waveform 0. 05 1 amp Amplitude against time 100 Hz frequency

Frequency: 500 -Hz Sine Wave 1. 0 Spectrum 0 Amplitude against frequency -1. 0

Frequency: 500 -Hz Sine Wave 1. 0 Spectrum 0 Amplitude against frequency -1. 0 0 Time (s) 0. 05 1 Waveform amp Amplitude against time frequency 100 500 frequency

Amplitude: 500 -Hz Sine Wave Spectrum Amplitude against frequency 0 1 0 Time (s)

Amplitude: 500 -Hz Sine Wave Spectrum Amplitude against frequency 0 1 0 Time (s) 0. 05 amp frequency 100 0 0 Time (s) 0. 05 500 frequency

Phase: 500 -Hz Sine Wave sine The amplitude spectrum does not show phase 1

Phase: 500 -Hz Sine Wave sine The amplitude spectrum does not show phase 1 amp cosine frequency 100 500 frequency

adding sine waves Spectrum of Sum 1 amp frequency 1 amp

adding sine waves Spectrum of Sum 1 amp frequency 1 amp

100 -Hz fundamental Complex Wave 5. 0 Spectrum Amplitude against frequency 0 -1. 7

100 -Hz fundamental Complex Wave 5. 0 Spectrum Amplitude against frequency 0 -1. 7 0 Time (s) Waveform 0. 05 1 amp Amplitude against time frequency 100 500 frequency

Frequency Adding nine sine waves Time 5 s Frequency Spectrogram

Frequency Adding nine sine waves Time 5 s Frequency Spectrogram

The linear vs log scales Linear • equal distances represent equal differences 0 100

The linear vs log scales Linear • equal distances represent equal differences 0 100 200 300 400 500 100 -1 200 0 400 1 800 2 1600 3 3200 4 Log • equal distances represent equal ratios e. g. Piano keyboard frequencies Octave = doubling of frequency basilar membrane has log repn of frequency

deci. Bel (d. B) scale Sound A is x d. B more intense than

deci. Bel (d. B) scale Sound A is x d. B more intense than sound B when: x = 10*log 10 (energy of A / energy of B) or x = 20*log 10 (amp of A / amp of B) So if A is 20 watts and B is 10 watts x = 10*log 10 (20/10) = 10*0. 3 = 3 d. B You can usually just hear a difference of 1 d. B (jnd)

Bandpass filtering (narrow) 5. 0 1 amp 0 -1. 7 0 Time (s) 0.

Bandpass filtering (narrow) 5. 0 1 amp 0 -1. 7 0 Time (s) 0. 05 frequency 100 500 frequency 1 amp 0 0 Time (s) 0. 05 frequency 100 500 frequency

Bandpass filtering (wide) 5. 0 1 amp 0 -1. 7 0 Time (s) frequency

Bandpass filtering (wide) 5. 0 1 amp 0 -1. 7 0 Time (s) frequency 100 0. 05 500 frequency 1 amp frequency 100

Beats Repetition rate is the difference in frequency between the two sine-wave components 1/3

Beats Repetition rate is the difference in frequency between the two sine-wave components 1/3 second 505 - 500 = 5 Hz 1 amp frequency 100 500 frequency

Beats Repetition rate is the difference in frequency between the two sine-wave components 1/100

Beats Repetition rate is the difference in frequency between the two sine-wave components 1/100 th second 500 - 400 = 100 Hz 1 amp frequency 100 400 500 frequency

Reponse of basilar membrane to sine waves Each point on the membrane acts like

Reponse of basilar membrane to sine waves Each point on the membrane acts like bandpass filter tuned to a different frequency: high freq at base, low at apex. Each point vibrates at frequency of pure tone (-> phase locking)

Excitation patterns (envelope of excitation) Basilar membrane excitation pattern is like a spectrum

Excitation patterns (envelope of excitation) Basilar membrane excitation pattern is like a spectrum

Auditory filter bandwidth (ERB)

Auditory filter bandwidth (ERB)

Excitation pattern of complex tone on bm

Excitation pattern of complex tone on bm

Measurement of auditory bandwidth with band-limited noise 1000 Hz 250 Hz Broadband Noise frequency

Measurement of auditory bandwidth with band-limited noise 1000 Hz 250 Hz Broadband Noise frequency 2000 Hz Amadeus

A gardening analogy

A gardening analogy

A gardening analogy

A gardening analogy

A gardening analogy Tone Noise Auditory bandwidth Noise bandwidth Detection mechanism

A gardening analogy Tone Noise Auditory bandwidth Noise bandwidth Detection mechanism

Wider auditory filter

Wider auditory filter

Auditory tuning curves Inner hair-cell damage Healthy ear

Auditory tuning curves Inner hair-cell damage Healthy ear

Outer-hair cell damage

Outer-hair cell damage

Human auditory bandwidth At 1 k. Hz the bandwidth is about 130 Hz; at

Human auditory bandwidth At 1 k. Hz the bandwidth is about 130 Hz; at 5 k. Hz the bandwidth is about 650 Hz. BW = freq / 8 roughly

Normal auditory non-linearities • Normal loudness growth (follows Weber’s Law, which is logarithmic, not

Normal auditory non-linearities • Normal loudness growth (follows Weber’s Law, which is logarithmic, not linear) • Combination tones • Two-tone suppression • Oto-acoustic emissions

Conductive vs Sensori-neural deafness Mostly a combination of OHC and IHC damage Becomes linear,

Conductive vs Sensori-neural deafness Mostly a combination of OHC and IHC damage Becomes linear, so No combination tones Or two-tone suppression

Symptoms of SNHL • Raised thresholds: helped by amplification • Wider bandwidths: no help

Symptoms of SNHL • Raised thresholds: helped by amplification • Wider bandwidths: no help possible • Recruitment (restricted dynamic range): partly helped by automatic gain controls in modern digital aids • Often accompanied by tinnitus

Normal vs Impaired Dynamic Range

Normal vs Impaired Dynamic Range