A comparison of acoustic features of speech of typically developing children and children with autism spectrum disorders

Transcription

1 * A comparison of acoustic features of speech of typically developing children and children with autism spectrum disorders Elena Lyakso, Olga Frolova, Aleksey Grigorev Saint Petersburg State University, Child Speech research group lyakso@gmail.com

2 The study aims The goal of this study is to find out the acoustic features specific for autism spectrum disorders (ASD) children vocalizations and speech 2

3 Method Participants of this study: 1. Typically developing children, aged 4-14 years (n=6) 2. Children with ASD (ICD - F84.), aged 5-14 years (n=25) 3

4 Childhood Autism Rating Scale - CARS (Shopler et al., 1986) 15 indicators Group-1 - development reversals; 1 - social interaction; 2 imitation; 3 - emotion; 4 body possession; 5 - object manipulation; 6 adaptation skills; 7 visual responses; 8 auditory responses; 9 tactile & gustatory responses; 1 - fear or anxiety; 11 verbal communication; 12 - non-verbal communication; 13 - activity level; 14 intellectual level 15 overall impression a mild form a severe form of autism Group -2 - the risk diagnosed at birth: ASD is a symptom of neurological diseases associated with brain disturbed the total scores, CARS scores, CARS ,5 3 2,5 2 1,5 1,5 group -1 group -2 Differences between groups 5- manipulation 6 - adaptation 12- non-verbal communication severe form mild form group-1 group -2 4

5 Speech & language TD normal speech words phrases complex phrases normal vocalizations - cry, vocalizations like-babbling; words; simple phrases ASD Repetitive same syllables, words, phrases special + personal utterances with non clear articulation Communication Articulation skills, prosody, grammar, pragmatics (used language inappropriately) - abnormalities TD: dialog with replica words & phrases ASD: 1. absence 2. dialog with simple replica: syllable, one words, simple phrases repetition 5

6 ASD: Characteristics of speech A. Characteristics of speech level B. Using the speech (wishes / reluctance) group 1 group group-1 group -2 child's, % vocalizations speech special speech special sp+speech wishes reluctance 6

7 Speech features: TD vs. ASD 3 experiments were conducted: Experiment- 1: Emotional speech; Experiment - 2: Spontaneous (spoken) speech; Experiment - 3: Words repetition

8 Experiment -1: Emotional speech TD: Modeling emotional state 1. Dialogue with the experimenter; 2. Playing with a standard set of toys 3. Repetition of words from a toy-parrot in a game store setting, and watching a cartoon - Masha and bear" from ipad and the retelling of the story, respectively places of recording were: at home, in laboratory and kindergarten 8

9 Experiment -1: Emotional speech ASD: Modeling emotional state 1. Swimming in the pool removal of stress 2. Show a standard set of the pictures and questions on the picture pictures description 3. playing with toys places of recording were: swimming pool & laboratory 9

10 Emotional state manifestation in child s speech: A. - recognition by adults % y 5 y 6-7 y A. TD Task -1: Discomfort Neutral speech samples 5 25 discomfort neutral comfort Сhild's emotional state Comfort speech / vocalizations, % discomfort neutral comfort Child's emotion group - 1 group - 2 B. ASD 1

11 Correct recognition ASD children emotional state A. words 5 Task -2: Fear Anger Sadness Calm Joy Surprise auditors, % 5 25 angry calm joy B. vocalizations auditors, % 25 fear anger calm joy 11

12 B - acoustical features F - pitch values, F3 F * * A: F average * TD ASD-1 ASD-2 F.Hz discomfort neutral comfort F * F3 * * F, Hz * B: Fmax-Fmin * * TD ASD-1 ASD-2 F max F3, Hz * * 1 5 * F min 1 discomf neutral comf discomfort neutral comfort F range = Fmax - Fmin * - p<.5; - p<.1; * p<.1- Mann-Whitney test 12

13 Intonation contour (prosody) ASD speech samples, % Falling - discomfort TD Rising - comfort Flat - neutral falling rising flat falling 4 y 5 y 6-7y!!!! For this samples of speech: Correlation between child emotion state and intonation are absent For vocalizations: Discomfort: falling, rising +falling + rising discomfotr confort neutral 13

14 TD: The words of children, reflecting different emotional states words, n discomfort comfort 4y 5y 6y 7y child's age Verb + adjective 14

15 Experiment - 2: spontaneous speech child dialogue with the experimenter (neutral theme as possible as)

16 Acoustic features of TD and ASD spontaneous child speech Pitch values Fo, Hz * * * * TD ASD- 1 ASD child age, years Pitch values of spontaneous speech of ASD children higher (p<.1) than pitch values of TD children, pitch values of ASD-1 children higher (p<.1) than in ASD-2. For all children with ASD voice and speech is characterized by high values of the pitch, abnormal spectrum, and well-marked high-frequency. 16

17 The intensity of the third formant En/E ASD-1 a * u i F F1 F2 F3 * En/E TD * F F1 F2 F3 a u i En/E ASD-2 a u i F F1 F2 F3-2 * -2 * * The distribution of the three first format s amplitudes normalized to the amplitude of pitch for vowels /a/, /u/, /i/ in spontaneous speech. Vertical axis En/E (normalized amplitude), horizontal axis F and formants (F1, F2, F3). A specific characteristic of the dynamic spectrum of the vowels of the ASD -1 child is the intensity of the third formant. The intensity of the vowel formants ASD -2 children was not significantly different from the TD corresponding data, except for F3/E for vowel /a/. 17

18 Experiment - 3: the words repetition the child repeated the words of the experimenter or parent (for ASD children)

19 Experiment -3: Repetition pith values (Fo), Hz Repetition vs. spontaneous child speech TD spontan TD repetit * ASD spont * ASD-1 spont * ASD-2 spont ASD repet ASD-1 repet ASD-2 repet The pitch median values in the stress vowels of ASD and TD children in twice speech tasks (spontaneous and repetition) Pitch values don't differ significantly in stress vowels from TD children's words in twice task. Pitch values of the ASD-1children was significantly higher (p<.1) than in the ASD-2 child s spontaneous speech. Pitch values variation (Fmax-Fmin) significantly higher in spontaneous speech of ASD-1 children than ASD -2 and TD children and in repetition words ASD-2 children were revealed. 19

20 The vowels formant triangles with apexes /a/, /u/, /i/ TD & ASD ASD-1 & ASD-2 F2, Hz /u/ /i/ A TD spont TD repet ASD spont ASD repet /a/ F2, Hz /i/ /u/ B /a/ ASD-1 spont ASD-1 repet ASD-2 spont ASD-2 repet F1, Hz F1, Hz 4 C 4 D 3 3 conv.un. 2 conv.un TD spont TD repet ASD spont ASD repet ASD-1 spont ASD-1 repet ASD-2 spont ASD-2 repet 2

21 Conclusion Differences between children with ASD and TD on the basis of higher values of pitch, pitch variability and formant characteristics of ASD children were revealed. In general, for all children with ASD voice and speech is characterized by high values of pitch and pitch variability. These acoustic features and well-marked high-frequency in spectrum were more clearly presented in the speech of the first group ASD children than the second group ASD children. The current results are one of the first steps toward developing speech based bio-markers for diagnosis of ASD. We believe that the acoustic features of speech of children with different neurological state are perspective for early diagnosis of developmental risk. 21

22 Thank for yours attention! 22

23 Experiment -3: the words repetition the child repeated the words of the experimenter or parent (for ASD children) Words to repeat: 1. Vowels /a/, /i/ and /u/ after the following consonants: /k/ and /d/ for /a/, /b/ and /g/ for /u/, /t / for /i/ / T'i / - тигр (tiger), ротик (mouth), / Ka / - кашка (clover), кар-кар (car-car), кап-кап (the drip-drip), кар (car), / Da / - дача (cottage), да (yes!), дай (give), / Bu / - буря (Storm), бух (wham!), буду (I will), могу (I can), арбуз (watermelon), бу-бу-бу (bu-bu-bu) / Gu / - губы (lips), в пургу (in a blizzard), гу-гу (gu-gu), гуси (geese) 2. Words with stressed vowels / a /, / i /, / u /

24 The context (/a/, /i/ and /u/ after the following consonants) influence on the characteristics of vowels in speech repeated was not significant for ASD and TD children