Speech Production and testing

Question 1

What is quick SIN?

Answer 1

A diagnostic ‘speech in noise’ test which tests the signal to noise ration loss
Ie how much do we need to boost signal so patient can hear over noise

Question 2

What is the quick SIN methodology?

Answer 2

6 sentences containing 5 key words
Presented at prerecorded SNR ratio
Decrease 5dB with each sentence
Normal to severely impaired performance in BGN
Defaults to 70dB but can be changed dependent on HL

Question 3

What are the quick SIN presentation levels?

Answer 3

identify PTA average:

Up to 45dB = 70dB
55. 80-85
65. 90-95
75. 100-105

Ie 25-30 above their PTA average

Question 4

What are the SNR results for quick SIN?

Answer 4

0-3. Normal hearing
4-7. Mild SNR loss mild issue in BGN
8-15 moderate SNR loss
15+ severe SNR loss

The greater the loss the better HA technology required
Record results on diagnostic page in CMS

Question 5

What are the types of speech issues that we can test using speech testing? (5)

Answer 5

Retro cochlea
Auditory ability - disparity between ears
Non organic hearing ie noise induced
Child testing
Hearing aid benefit

Question 6

Name the different speech tests for adults?

Answer 6

Speech recognition threshold test
Speech discrimination tests
Arthur Boothroyd
Acceptable noise level
Audibility index audiology
Speech shaped noise
HH consultation - speech test

Question 7

What is the speech recognition threshold test?

Answer 7

Spondee words 2 equal words together ie eardrum airplane schoolboy etc

Word presented patient repeats
If correct increase by 10dB with another word
If wrong decrease by 5dB until heard

Need 3 out of 6 for each ear
Should be within 10dB of PTA average

Question 8

Describe the speech discrimination test.

Answer 8

Phonetically balanced words used: equally difficult, common use and monosyllabic.
Called the ‘fry’ list.
Do the results match PTA?
Can identify asymmetrical loss

Question 9

What is the Arthur Boothroyd test?

Answer 9

Uses 3 letter consonant-vowel-consonant words for adults add older children.

10 words - 1 point given for each letter correct. Maximum 30 points available.

Patient repeats what they think has being said.

Question 10

Describe the acceptance noise level (ANL) test.

Answer 10

Field free - recorded voice through loudspeakers.
A speech passage (story) and noise

ANL = MCL - BNL

Patient listens to story, volume set at their MCL.
Gradually introduce noise to the story until patient reports maximum Levi they could listen to story and still understand speech (background noise level BNL)

The lower the score the better chance of a good fitting.

Question 11

Explain the audibility index audiogram.

Answer 11

COUNT THE DOTS!!
A speech intelligibility index we count all the dots BELOW the hearing threshold (each ear) to determine score.
Dots strategically placed within specific points of speech bubble/banana

Question 12

What is the speech shaped noise test?

Answer 12

White noise used to block cochlea. White noise is same intensity across all frequencies.
With SSN patient will capture some sounds as intensity is WEIGHTED across frequencies to mimic speech sounds.

Question 13

HH speech test?

Answer 13

Unaided-field free-live voice
List of balanced words
Max 10 words
Patient closes eyes & repeats what they hear
Score out of 10

Question 14

What is speech?

Answer 14

A process of uttering ARTICULATED sounds.
A complete feedback process in which hearing, perception and information processing in the nervous system and the brain are also involved.

Question 15

Explain harmonics add timbre

Answer 15

A harmonic is a sound wave that has a frequency that is a multiple (whole number) of a fundamental tone - the lowest frequency of that sound.

2nd harmonic (f2) freq twice that of fundamental
4th harmonic (f4) four times that of fundamental frequency.

Harmonics and overtones to a wave that make possible to distinguish between sounds and instruments - known as timbre

Question 16

What is a Formant?

Answer 16

It is the acoustical energy concentrated around a particular frequency in speech.

Vocal cords set up secondary vibrations (harmonics) which enhance some frequencies more than others.
These enhanced frequencies are Formants.

Question 17

Why is noise often heard as low frequency?

Answer 17

Noise contains a mix of frequencies but high frequencies do not travel well due to being a shorter wavelength, therefore, lose their energy and thereby intensity sooner.

Question 18

What is frequency analysis?

Answer 18

This breaks a frequency down into its component parts ie the different frequencies that make up a sound - the more frequencies the more complex the sound.

Question 19

What is the SNR? When does it become positive, negative or equal?

Answer 19

Signal to noise ratio.
It is the balance between the signal strength and the BGN strength.

To hear speech in noise clearly the signal needs to be equal or more than the noise signal depending on the HL.

It is a DECIBEL VALUE.

If signal is stronger than noise it is POSITIVE.
If it is weaker than noise it is NEGATIVE.

0dB SNR is where they are equal.

Conductive loss needs 0dB SNR as long as the SNR is above their hearing threshold. Ie a 30dB conductive loss would require a 0dB SNR to be at 30dB.
A SN loss typically needs SNR to be POSITIVE PLUS 10dB.

Question 20

What is the Lombard effect?

Answer 20

Simply that the voice will naturally amplify with the intensity of noise ie to ensure will be heard over the noise.

Question 21

What are the 4 key elements of speech production? Explain the function of each.

Answer 21

RESPIRATION - provides energy for sound by using air pressure (exhalation) from the lungs.
PHONATION - vibrating vocal chords via rapid opening and closing of the chords.
RESONANCE - air pressure and added vibrations now resonate in the vocal tract (air cavity above the Chords that extends to the lips)
ARTICULATION - lips, tongue, jaw and soft palate of mouth are used differently to make different sounds - the SPEECH SIGNAL.

Question 22

Describe the journey of the speech signal from lungs to lips.

Answer 22

As we exhale this pushes the air out with more force which travels up the trachea to the larynx (voice box) causing vocal chords to rapidly open and close causing vibrations.

Vocal cords elongate - longer produces low frequency and shorter produces high frequency. It is the length of the vocal cord that determines the RESONANT frequency.

These vibrations then enter the vocal tract air cavity where they resonate.
Male vocal tract is low freq around 120Hz and female/child 220Hz higher frequency.

The articulators then shape the sound to produce speech.

Question 23

What are the most effective resonators?

Answer 23

1st Pharynx
2nd mouth
3rd nose

Question 24

Speech sound is divided into two types - what are they and what differentiates between them?

Answer 24

VOICED: produced in larynx by means of expelled air from lungs and VIBRATED by the vocal cords.

VOICELESS: exactly the same but they are NOT VIBRATED by the vocal cords.

Question 25

What are vowel sounds and how do they differ from consonants?

Answer 25

Vowels is air from the lungs not blocked by the mouth or throat (consonants are)
Vowels are always VOICED.

Question 26

What is the average speech intensity?
Relate this to the inverted square law.

Answer 26

Approximately 65 dB at a distance of 1 metre. However, fluctuation of 12dB between quietest and loudest sounds.

Relate: the intensity to the inverted square law ie 6dB loss as you double the distance.
Intensity is always proportional to the distance squared.

Question 27

How do recognise speech sounds?

Answer 27

Speech is a series of sounds made into words.
Phonemes are individual sounds, the smallest unit of speech.
Then syllables
Then words
Then sentences

Ie PLANET
6 phonemes
2 syllables
1 word
0 sentence

Question 28

What are the 3 types of consonants? Describe them.

Answer 28

NASAL: M, N, NG mouth closed, sound out of nasal cavity.

PLOSIVES: vocal tract blocked somewhere then air pressure released.
Voiced: P K T voiceless: B D G

FRICATIVES: voice Tract narrow, air forced through.
Voiced S F voiceless Z V

Question 29

Which sounds do we hear better? Why?

Answer 29

Vowels are low frequency soo heard well whilst consonants are high frequency soo hard less well.

BGN is generally low frequency so this increases the chances of the quieter, high frequency consonants not being heard.

REMEMBER: low equals volume/Power high frequency is clarity.

Question 30

What aspects of hearing loss will affect speech production?

Answer 30

Type and degree of HL
severity of the loss
Nature of onset and it’s progress
Duration elapsed before HA’s.

Question 31

What are the key stages in human speech development?

Answer 31

4-6 months babbling
7-11 months tuneful babbling
12, months couple of words
18-24 mths 2 words together
2 years building lines ‘me do it’
3 years 5-7 word sentence
3 plus using ‘ing’ words plurals etc when speech issues can be identified.