What makes a charismatic speaker? A computer-based acoustic-prosodic analysis of Steve Jobs tone of voice

Source: @niebuhr2016a

Highlights 💡

Hinzugefügt am 2022-10-10

acoustic profile of Steve Jobs (p. 1)

4000 syllables and 12,000 individual speech sounds from his two (p. 1)

presentations: the introductions of the iPhone 4 and the iPad 2 (p. 1)

stands out against our reference samples in almost all key features of charisma, including melody, loudness, tempo and fluency (p. 1)

quantitative differences when addressing customers and investors (p. 1)

Strange (2007:434) (p. 1)

researchers today successively converge to the conclusion that being a charismatic leader necessarily entails being a charismatic speaker (p. 1)

speech is even one of the key ingredients of charismatic leadership (p. 1)

Signorello, interviewed by Hogenboom, BBC, 2014 (p. 1)

objective acoustic measurements (p. 2)

multi-parametric quantitative approach (p. 2)

we shed some initial light on the widely open question if and how charismatic speech differs for different audience groups or in different situations (p. 2)

constitute the “tone of voice” as ‘melodic features’ (p. 2)

speaking rate, loudness, speech rhythm, speech melody, fluency, and emphatic accentuation (p. 2)

  • Elizabeth Holmes, the way they do it for Jobs in this section ✅ 2022-10-31

  • [b] AM2 Charisma: Im Gegensatz zu Steve Jobs ist Holmes tief gefallen → das sollte Bewertungen ihres Charisma für diejenigen, welche die Geschichte aktiv verfolgt haben/denen die Erinnerung an den Schwindel noch im Kopf herumschwebt, ziemlich weit nach unten ziehen

  • AM2 Charisma: War es schwer für mich, genug Redematerial zu finden? ✅ 2022-10-31

as these speeches were product presentations whose content is clearly and constantly structured, we were able to reliably operationalize audience groups (p. 2)

Charismatic speakers are rare, particularly those who are undoubtedly and across the board perceived to be charismatic. (p. 2)

In this way, our “connected” multimedia society weakens traditional hierarchies based on dominance and authority; and, at the same time, our digital world is fast, which makes it more difficult for individuals to stand out from the crowd and win over people in a short time. Under these conditions, charisma and charismatic speech have become an ever more important – effective and efficient – factor for success and leadership. (p. 3)

Harrington (2010:82) (p. 3)

Harrington & Cassidy, 2012 (p. 3)

first investigations of charisma were concerned with strategies of effective leading and characterized by pure a posteriori descriptions. (p. 3)

inherent and categorical personal trait (p. 3)

communicate effectively and persuasively (p. 3)

attract and retain followers (p. 3)

gain authority without any institutional power (p. 3)

situations in which charisma is used and perceived (p. 3)

embedded. (p. 3)

learnable and shapeable skill (p. 3)

intensive training can make speakers gain conscious control over their tone of voice (p. 3)

charisma can be more or less pronounced, and it can change over time (p. 3)

some speakers are better than other speakers in producing charismatic speech (p. 3)

Sørensen (2013: 24) states, “language forms the basis of charisma” (p. 3)

fluency,…energy, and voice tone variety (p. 3)

Delivery comprises all nonverbal aspects of a message that range from body-language features to the melodic features of the speaker’s tone of voice (Awamleh & Gardner, 1999; Holladay & Coombs, 1993, 1994) and its emotional and motivational effects on followers (Shamir et al., 1994). Delivery is responsible for yielding credibility (Holladay & Coombs, 1993, 1994). Credibility contributes to persuasion (Sørensen, 2013), and persuasion, in turn, plays a key role in creating charisma (Awamleh & Gardner, 1999). (p. 4)

Among the few acoustic-phonetic studies on the production and perception of melodic features are Rosenberg and Hirschberg (2005; 2009), Signorello, D’Errico, Poggi, and Demolin (2012), Signorello, D’Errico, Poggi, Demolin, and Mairano (2012) and D’Errico et al. (2013). (p. 4)

Charisma means leading without being dominant (p. 4)

sections of official and thus strongly conventionalized speeches (p. 5)

single genre (p. 5)

constant speaking situation, content structure, and audience composition (p. 5)

Sections (2) and (3) are the sections we extracted (p. 5)

From each type of section, we pseudo-randomly selected about 11 min of speech, 5e6 min from the iPhone 4 and another 5e6 min from the iPad 2 presentation, which gave us a total amount of 22 min of speech data for acoustic analyses, consisting of about 4000 syllables and 12,000 individual speech sounds. (p. 5)

disregarded sections with insufficient audio quality (p. 5)

due to partial masking of the speech signal by music, applause, or noise (p. 5)

the compression rate (160 kbps) (p. 6)

separating Steve Jobs’ parts of speech from other types of speech signals (p. 6)

intended silent pauses (annotated as ) (p. 6)

disfluencies due to silent hesitations (hes:s) (p. 6)

filled pauses (hes:e, e.g., “err”, “um” as well as repetitions and lengthened syllables, cf. Duez, 1982) (p. 6)

interruptions due to breathing (hes:a) (p. 6)

orthographically transliterated (p. 6)

Beginnings and ends of parts of speech were determined with reference to the well-established “breath-group” (p. 6)

312 parts of speech in the customer-related (p. 6)

77 parts of speech in the investor-related (p. 6)

vowel onsets and offsets were automatically determined and annotated in PRAAT by mapping the orthographic transliteration onto the loudness contour (p. 6)

Mean pitch level of each part of speech in terms of the signal’s fundamental frequency (F0, in Hz). (p. 6)

Pitch variation in each part of speech in terms of the standard deviation (in Hz) and the range (in semitones, st) (p. 6)

Mean loudness level of each part of speech in terms of rootmean-square intensity (RMS, in dB) (p. 6)

Loudness variation in each part of speech in terms of the standard deviation (in dB) (p. 6)

Duration of each part of speech (in seconds (p. 6)

Duration of each intended silent pause, disfluency, and interruption due to breathing (p. 6)

Speaking rate of each part of speech in terms of syllables per second (syl/s) (p. 6)

Speech rhythm characteristics of each part of speech in terms of the vowel-based parameters %V and VarcoV (p. 6)

Total number of intended silent pauses, disfluencies, and interruptions due to breathing (p. 6)

Total number of emphatically accented words (p. 6)

  • Research %V and VarcoV

  • Search for Praat scripts

deliberately mixed up factors such as the recording condition, the speakers’ tasks, environmental conditions, and the linguistic background (p. 7)

T-tests (p. 7)

  • Christ, did they even publish this compiled dataset somewhere? Should I ask?

dot size is an index of pitch variability, produced by dividing the means values by their standard deviations. This variability index is also known as Pitch Dynamism Quotient (PDQ) (p. 7)

  • Forschungsergebnisse

‘Accent chains (p. 9)

’word repetitions’ (p. 9)

’Reinforcement (p. 9)

’Positive intensification’ (p. 9)

Steve Jobs makes use of all the melodic features of charisma that were identified – for political speeches – in previous analysis and experiments (p. 10)

his product presentations represent a highly skilled implementation of speech-based charisma features (p. 10)

Levine et al., 2010) (p. 11)

charisma is not created by constant changes in one or two acoustic parameters, but relies on an entire bundle of melodic features (p. 11)

the frequency code’ (Gussenhoven, 2002; Ohala, 1994) (p. 13)

a charismatic speaker, by definition, does not need to be dominant in order to attract and retain followers (p. 13)

uncontrolled differences in recording gains, loudness normalization procedures or measurement methods between Steve Jobs’ data and the reference sample (p. 13)

cultural or social dimensions of charismatic speech (p. 13)

language level (p. 13)

whose melodic systems strongly differ in terms of the communicative functions (p. 13)

personality of the presenter and the industry he or she is working in (p. 13)

gender (p. 13)

wording, gestures, outer appearance, clothes, situation, social status, design of supporting presentation slides, familiarity, medium (p. 13)

So far, all studies on the acoustics of charisma dealt with public speeches. (p. 13)

pedagogy. (p. 14)

Borský, M., & Poll ak, P. (2015). Analysis and automatic recognition of compressed speech. In O. Niebuhr, & R. Skarnitzl (Eds.), Tackling the complexity in speech (pp. 205e221). Prague: Charles University Press. (p. 15)

Gussenhoven, C. (2002). Intonation and interpretation: phonetics and phonology. In Proc. 1st International Conference of Speech Prosody, Aix-en-Provence, France (pp. 47e57). (p. 15)

Signorello, R., D’Errico, F., Poggi, I., & Demolin, D. (2012, September). How charisma is perceived from speech. A multidimensional approach. In Proc. ASE/IEEE international conference on social computing, Amsterdam, Netherlands (pp. 435e440). (p. 16)

Sørensen, L. S. (2013). How to grow an apple: Did Steve Jobs speak Apple to success? An analysis of Steve Jobs’ rhetorical and linguistic development in relation to Apple’s organizational performance. Masterthesis. Aalborg University. (p. 16)

Strange, K. D. (2007). Examining effective technology project leadership traits and behaviors. Computers in Human Behavior, 23, 424e462. (p. 16)