Sound Follows Function

wew

Sound follows function Sound communication and the relevance of timbre. Lecturer: Rainer Hirt & Kai Bronner (audio-branding.de)

wew

01 Acoustic signs and its functions 02 Overview ongoing Functional-Timbre-Study (FTS)

wew

01 Acoustic signs and its functions 02 Overview ongoing Functional-Timbre-Study (FTS)

wew

01 Acoustic signs and its functions

wew

A. Basics

Signs?

A. Basics

B. Study

C. Results

Signs! A sign is a stimulus pattern that has a meaning. A sign stands for something else.

A. Basics

B. Study

C. Results

Semiotics: Theory of signs Index An „Index“ is defined by some sensory feature (something directly visible, audible, smellable, etc) that is connected to it.

A. Basics

B. Study

C. Results

Semiotics: Theory of signs Icon An icon is a pattern that physically resembles what it „stands for“.

A. Basics

B. Study

C. Results

Semiotics: Theory of signs Symbol Symbols are arbitrary and unmotivated, reliant on conventional usage to determine meaning.

A. Basics

B. Study

C. Results

Acoustic signs?

A. Basics

B. Study

C. Results

Acoustic signs! Index Icon Symbol Index becomes Icon Symbol becomes Icon

A. Basics

B. Study

C. Results

Applications of acoustic signs Cultural

Auditory Display

A. Basics

Corporate Sound

Auditory-User-Interface

Audification

B. Study

C. Results

Basic design-approaches of acoustic signs: Auditory icons Auditory symbols Auditory symcons (Mixed) A. Basics

B. Study

C. Results

wew

B. Empirical study

Ambition & research questions 1. Appropriateness of Auditory icons/symbols and symcons for the transfer of information-functions: –> Advice –> Warning –> Emergency 2. Effects of the mixture of icon and symbol („auditory symcon“) 3. Advantages of acoustic signs for „public-user-interfaces“

A. Basics

B. Study

C. Results

Analysis of a situation –

Talks with manfactures and customers (Höft-Wessel, ICA, DB)

–

Secondary literature

–

Preliminary study (semantic field-analysis)

A. Basics

B. Study

C. Results

Study object Ticket machine

A. Basics

B. Study

C. Results

Study design Study method: Descriptive study Survey sample size : N=22 Age average: 35 years Women/Men: 10/12

A. Basics

B. Study

C. Results

Lab set-up > Lectern (Terminal-simulation) > Table-PC (Touchpad) > Headphone

A. Basics

B. Study

C. Results

Stimuli

c c i i n l o o c i b c n m m ico sy sy

Advice: „Ticket ready for collection“

1

2

3

Advice: „Money ready for collection “

1

2

3

Warning: „Timeout“

1

2

3

Emergency: Deterrence

1

2

3

Context-Soundscape (Trainstation)

A. Basics

B. Study

C. Results

Evaluation methods Semantic differential

A. Basics

B. Study

C. Results

Evaluation methods Semantic differential Categorical scaling

A. Basics

B. Study

C. Results

Evaluation methods Semantic differential Categorical scaling Response time measurement

A. Basics

B. Study

C. Results

Evaluation methods Semantic differential Categorical scaling Response time measurement Participant observation

A. Basics

B. Study

C. Results

Evaluation methods Semantic differential Categorical scaling Response time measurement Participant observation Association analysis A. Basics

B. Study

C. Results

wew

C. Results

Ambition & research questions 1. Appropriateness of Auditory icons/symbols and symcons for the transfer of information-functions: –> Advice –> Warning –> Emergency

2. Effects of the mixture of icon and symbol („auditory symcon“)

A. Basics

B. Study

C. Results

Ambition & research questions Auditory Icon

Auditory Symbol

Auditory Symcon

Advice Warning Emergency Funtion(intention) partially transferred Funtion(intention) transferred Funtion(intention) not transferred or false interpretation

A. Basics

B. Study

C. Results

Ambition & research questions 3. Advantages of acoustic signs for „public-user-interfaces“

A. Basics

B. Study

C. Results

Ambition & research questions 3. Advantages of acoustic signs for „public-user-interfaces“ The meaning of function of acoustic signs is comprehended intuitively.

A. Basics

B. Study

C. Results

Further investigations The study reveals amongst others, that further investigation on the acoustical parameter „timbre“ is required. In the literature there is nearly no advice for selecting timbre according an information-function. Thus, an ongoing study that addresses the perception of timbre is topic of the further presentation.

A. Basics

B. Study

C. Results

wew

02 Timbre and function

A. Timbre: definition & properties B. Sound perception C. Study design

A. Timbre: definition & properties

B. Sound perception

C. Study design

wew

A. Timbre: definition & properties

Timbre – what? Definition Definition (American Standards Association 1960) Timbre is that attribute of auditory sensation in terms of which a listener can judge that two sounds similarly presented and having the same loudness and pitch are dissimilar.

A

A. Timbre: definition & properties

B

B. Sound perception

C. Study design

Timbre properties Verbal descriptions overtone constellation: bright, dark, mellow, hollow, pure noise content: raspy, breathy, hoarse attack: smooth, abrupt, sharp, gentle, easing Relations pitch - frequency loudness - amplitude timbre - multiple interacting acoustic factors

A. Timbre: definition & properties

B. Sound perception

C. Study design

Timbre properties H. Fletcher (1934) Experiments show that a simple one-to-one relationship does not exist between the two sets { loudness, pitch, timbre } and { sound intensity, fundamental frequency and overtone structure }

sound intensity fundamental frequency overtone structure

A. Timbre: definition & properties

B. Sound perception

loudness pitch timbre

C. Study design

Timbre properties More than one dimension: multidimensionality Erickson (1975) Clearly timbre is a multidimensional stimulus: it cannot be correlated with any single physical dimension. Plomp (1982) Sounds cannot be ordered on a single scale with respect to timbre. Timbre is a multidimensional attribute of the perception of sounds.

A. Timbre: definition & properties

B. Sound perception

C. Study design

Multi-Dimensionality-Scaling MDS MDS studies reveal perceptual dimensions correlated with acoustic parameters corresponding to spectral, temporal and spectrotemporal properties of the sound events. – experiments with paired sound stimuli – rating of similarity – axes corresponding to timbre‘s main perceptual dimensions –> timbre space

A. Timbre: definition & properties

B. Sound perception

C. Study design

Dimensions Dimension I = brightness Low brightness : french horn and cello High brightness : oboe, muted trombone

Dimension I

Dimension II = spectral flux High synchronicity and low fluctuation : clarinet, saxophone Low synchronicity and high fluctuation : flute, violoncello

Dimension III = attack quality More transients : strings, flute Fewer transients : brass, bassoon

A. Timbre: definition & properties

B. Sound perception

Dimension II

sion n e m i D

C. Study design

III

Timbre dimensions and related parameters Across studies, several acoustic/physical parameters corresponding to timbre dimensions were found: – Spectral centroid

– Spectral flux

– Spectral deviation

– Pitch strength

– Spectral density

– Attack synchronity

– Attack time

– Attack centroid

– Decay time

– Noisiness

–Amplitude envelope

...

A. Timbre: definition & properties

B. Sound perception

C. Study design

wew

B. Sound perception

Timbre and psychoacoustic properties Interview Dr. Gerhard Thoma, chief sounddesigner BMW: (http://www.goethe.de/ges/wrt/dos/aut/sou/de2169434.htm , 10/2007)

(...) There are sets of regression equations that help calculating/predicting a desired sound perception, for example a sporty sound: a x sharpness + b x roughness + c x fluctuation strength, … = sportiness.!! About 80 % of people will evaluate the sound in terms of sportiness. But:„acid test“in driving situation (context) = proof of validity!(...) A. Timbre: definition & properties

B. Sound perception

C. Study design

Psychoacoustic properties and cognitive/affective evaluation –

simple psychoacoustic parameters only refer to fixed relations of perception and instantaneous stimuli (Haverkamp 2007)

–

psychoacoustics are only capturing parts of human sound perception, largely excluding important affective + cognitive evaluation (Vjästfall/Kleiner 2002)

–

any endeavour to evaluate (product) sounds by psychoacoustics solely will fail in the end (Blauert/Jekosch 1997)

A. Timbre: definition & properties

B. Sound perception

C. Study design

Sound perception: target groups/context – –

individual differences of preferences of particular groups of people for qualitatively different product sounds no simple „universal“ manipulation that will have the same effect on the sound quality of products (Spence & Zampini 2006)

–> target groups (age, cultural background, milieus/lifestyle) have to be taken into account –> it is important to consider the kind/type of product = context

A. Timbre: definition & properties

B. Sound perception

C. Study design

Emotional evaluation and functional content Activation tense

Model of affective circumplex (Russell, 1980)

alert

Emergency nervous

excited elated

stressed Warning Unpleasant

upset

happy Confirmation

Advice

contented

sad

serene

depressed bored

relaxed Deactivation

A. Timbre: definition & properties

B. Sound perception

C. Study design

Pleasant

Sound perception: evaluation of sound (timbre) To be taken in account: –

psychoacoustic properties of sound

–

learned associations

–

cultural background and personal experiences (target group)

–

context: mood of subjects, expectations, situation

–

influence from other modalities (vision, touch,...)

A. Timbre: definition & properties

B. Sound perception

C. Study design

wew

C. Study design

Functional Timbre Study („sound-colour the Gestalt“) VARIABLES (TIGA) Independent: Timbre T Gestalt G [pitch contour, rhythm, tone duration] Dependent: Information-function I [function, e.g. emergency, warning, advice/confirmation] Attribution A [emotional/affective evaluation: warm, harsh, sharp,...]

A. Timbre: definition & properties

B. Sound perception

C. Study design

Sound-colour the Gestalt Approach: sound-colour T the Gestalt G and examine evaluations of subjects regarding information-function I and attribution A

A. Timbre: definition & properties

B. Sound perception

C. Study design

Example of stimuli Same Gestalt, different timbre

A

A. Timbre: definition & properties

B

B. Sound perception

C. Study design

Study Design Step 1: free association task 3 Gestalt x 8 Timbre = 24 sound-stimuli are evaluated by subjects –> pool of attributes Step 2: rating of stimuli Rating of 24 stimuli on Likert scales of attributes obtained from step 1 + assessment of given information function Step 3: analysis of results Principal-Component-Analysis, Generalized-Linear-Models. Examination of correlations between acoustic properties, rated attributes, information function. –> Findings and results serve as a basis for further investigations! A. Timbre: definition & properties

B. Sound perception

C. Study design

Study Design Timbre selection Timbre differing in parameter values („timbre dimensions“)

bright

spectral content: temporal content: attack/cutoff:

static hard

dull/dim dynamic

soft

A. Timbre: definition & properties

hard

static soft

hard

B. Sound perception

dynamic soft

hard

C. Study design

soft

Study Design Gestalt creation Typical forms of sound-gestalt (learned associations) descending major third of a doorbell („ding dong“) (association: welcome) wailing siren, in germany a fourth: a-d a-d (police or ambulance) (association: emergency - alert) Scherer/Osinsky 1977 Tonality: major mode: indicative of pleasantness and happiness minor mode: disgust and anger Rhythm:

rhythmic: more active, fearful, surprised nonrhythmic: boredom

Activity = Fast tempo, high pitch level, many harmonics, large pitch variation, sharp envelope, small amplitude variation A. Timbre: definition & properties

B. Sound perception

C. Study design

Fields of expertise/collaboration Relevant fields of expertise: music psychology, sound-/acoustic-communication, sounddesign, sound-/music-computing, statistical methods –> interdisciplinary team

We are not the only ones: Product Sound Design Group TU Delft Music Technology Group Barcelona Music, Mind and Machine Group (MIT Media Lab , USA) Music Information Retrieval Community –> exchange of findings, knowledge + possible collaborations

A. Timbre: definition & properties

B. Sound perception

C. Study design

What is the use?? Product Sounddesign Audio-Branding AUI (Auditory User Interfaces) Music Information Retrieval MIR (music catalogues, sound-databases) Mpeg7-Classification Electronic Music Distribution „Perceptual Software-Synthesizers“ ...

A. Timbre: definition & properties

B. Sound perception

C. Study design

wew

Thank you!