Slow Speech Rate Research Articles

Speech Rate Adjustments in Conversations With an Amazon Alexa Socialbot

This paper investigates users’ speech rate adjustments during conversations with an Amazon Alexa socialbot in response to situational (in-lab vs. at-home) and communicative (ASR comprehension errors) factors. We collected user interaction studies and measured speech rate at each turn in the conversation and in baseline productions (collected prior to the interaction). Overall, we find that users slow their speech rate when talking to the bot, relative to their pre-interaction productions, consistent with hyperarticulation. Speakers use an even slower speech rate in the in-lab setting (relative to at-home). We also see evidence for turn-level entrainment: the user follows the directionality of Alexa’s changes in rate in the immediately preceding turn. Yet, we do not see differences in hyperarticulation or entrainment in response to ASR errors, or on the basis of user ratings of the interaction. Overall, this work has implications for human-computer interaction and theories of linguistic adaptation and entrainment.

Gesture, prosody and verbal content in non-fluent aphasic speech

Abstract Non-fluent aphasia is characterized by frequent word search and a much slower speech rate than non-aphasic speech. For patients with this type of aphasia, communication with those around them is therefore made difficult and is often severely impaired. One of the therapeutic proposals to improve the quality of life of these patients is to re-educate them with more multimodal alternatives. This of course assumes that gestures represent possible alternative means of communication for patients, and that their gestures are not affected in the same way as their speech. This article therefore proposes to study the gestures of 4 aphasic people and to compare them to the gestures performed by non-aphasic people, but also to establish correspondences between those gestures, intonation contours and the way people with aphasia develop their discourse. Results show that although gesture rate is not different in the two groups of participants, the gesture-to-speech ratio is higher for people with aphasia (PWA) than for non-aphasic people (NAP). Considering the fact that PWA also gesture more than NAP during silent pauses, which are longer but not more frequent than in NAP’s speech, and the fact that their gestures coincide less often with a lexical word, we believe that PWA use their gestures as compensation strategies for deficient speech. Yet, their speech impairment is also reflected in their gesturing: more gestures are prepared but abandoned before the stroke in this group and pre-stroke holds are longer, which means that PWA hold their gestures in the hope that they will better coincide with the word they are supposed to accompany and which takes more time to be uttered than in non-pathological speech. Their gestures are also less linked to each other than in the NAP group which goes hand in hand with the fact that they tend to utter independent syntactic phrases with no cohesive marker between them. This is also reflected in their less frequent use of flat and rising tones in intonation, which generally indicate that two sentence parts are dependent one upon the other, as well as their less frequent use of gestures showing discourse organization. In terms of gesture types, the PWA in this study perform many rhythmic beats and rely much on conventional gestures to compensate for their speech impairment rather than on their own creativity. Globally, this means that if multimodal therapies may benefit PWA to improve their communication with other people, speech therapists nevertheless need to be aware that life-long habits of gesture-speech alignment and synchronization may not be so easy to overcome for patients.

Impact of Cognitive Impairment and Dysarthria on Spoken Language in Multiple Sclerosis.

To investigate the impact of cognitive impairment on spoken language produced by speakers with multiple sclerosis (MS) with and without dysarthria. Sixty speakers comprised operationally defined groups. Speakers produced a spontaneous speech sample to obtain speech timing measures of speech rate, articulation rate, and silent pause frequency and duration. Twenty listeners judged the overall perceptual severity of the samples using a visual analog scale that ranged from no impairment to severe impairment (speech severity). A 2 × 2 factorial design examined main and interaction effects of dysarthria and cognitive impairment on speech timing measures and speech severity in individuals with MS. Each speaker group with MS was further compared to a healthy control group. Exploratory regression analyses examined relationships between cognitive and biopsychosocial variables and speech timing measures and perceptual judgments of speech severity, for speakers with MS. Speech timing was significantly slower for speakers with dysarthria compared to speakers with MS without dysarthria. Silent pause durations also significantly differed for speakers with both dysarthria and cognitive impairment compared to MS speakers without either impairment. Significant interactions between dysarthria and cognitive factors revealed comorbid dysarthria and cognitive impairment contributed to slowed speech rates in MS, whereas dysarthria alone impacted perceptual judgments of speech severity. Speech severity was strongly related to pause duration. The findings suggest the nature in which dysarthria and cognitive symptoms manifest in objective, acoustic measures of speech timing and perceptual judgments of severity is complex.

Referential context and executive functioning influence children's resolution of syntactic ambiguity.

Classic studies reveal two striking differences between preschoolers and adults in online sentence comprehension. Adults (a) recruit referential context cues to guide syntactic parsing, interpreting an ambiguous phrase as a modifier if a modifier is needed to single out the intended referent among multiple options, and (b) use late-arriving information to recover from misinterpretation. Five-year-olds fail on both counts, appearing insensitive to the referential context and often failing to recover from parsing errors (Trueswell, Sekerina, Hill, & Logrip, 1999). But other findings suggest that 5-year-olds show delayed rather than absent sensitivity to the referential context, and that individual differences in executive functioning predict children's ability to recover from garden-path errors. In 2 experiments, we built on these findings, focusing on whether children recruit referential-context cues if given time to do so. Children heard temporarily ambiguous instructions (e.g., Put the frog on the pond into the tent), while we monitored their eye-gaze and actions. We used a slow speech rate, and manipulated referential context between rather than within subjects, to give children time to bring referential context cues into play. Across experiments, eye-movement and action analyses revealed emerging sensitivity to the referential context. Moreover, error rates and eye-movement patterns indicating failures to revise were predicted by individual differences in executive function (scores in Simon Says and Flanker tasks). These data suggest that children, like adults, use referential context information in syntactic processing under some circumstances; the findings are also consistent with a role for domain-general executive function in resolution of syntactic ambiguity. (PsycInfo Database Record (c) 2020 APA, all rights reserved).

Editors' note: Motor speech patterns in Huntington disease

In the article “Motor speech patterns in Huntington disease,” Drs. Diehl et al. identified 4 distinct presentations of dysarthria in patients with Huntington disease (HD) based on the evaluation of audio recordings of 48 patients by 6 trained raters. The authors postulated that these distinct presentations may be due to divergent pathologic processes. In response, Drs. Rusz and Tykalova argue that the findings of a patient subgroup characterized by an abnormally fast speaking rate was at odds with prior evidence indicating slow speech rate in HD. They note that patients with a bradykinetic phenotype have previously been shown to have a slower speaking rate than those with a choreatic phenotype, challenging the authors' explanation of fast speaking rate as being related to hypokinetic dysarthria. Besides noting the absence of a control group of healthy patients for comparison, they also contend that the remaining 3 HD dysarthric subgroups may simply reflect overall disease progression. Responding to these comments, the authors cite prior work reporting a higher rate of speech in patients with HD and argue that the studies cited by Drs. Rusz and Tykalova relied on acoustic measures that may only weakly correlate with perceptual ratings of dysarthric speech. They note that the trained raters were asked to rate speech relative to what they considered normal, and that their study identified CAG repeats and Unified HD Rating Scale motor scores as differing significantly across speech subgroups. This exchange highlights potential challenges with reconciling acoustic-based and rater-based phenotyping of speech patterns in neurologic disease. In the article “Motor speech patterns in Huntington disease,” Drs. Diehl et al. identified 4 distinct presentations of dysarthria in patients with Huntington disease (HD) based on the evaluation of audio recordings of 48 patients by 6 trained raters. The authors postulated that these distinct presentations may be due to divergent pathologic processes. In response, Drs. Rusz and Tykalova argue that the findings of a patient subgroup characterized by an abnormally fast speaking rate was at odds with prior evidence indicating slow speech rate in HD. They note that patients with a bradykinetic phenotype have previously been shown to have a slower speaking rate than those with a choreatic phenotype, challenging the authors' explanation of fast speaking rate as being related to hypokinetic dysarthria. Besides noting the absence of a control group of healthy patients for comparison, they also contend that the remaining 3 HD dysarthric subgroups may simply reflect overall disease progression. Responding to these comments, the authors cite prior work reporting a higher rate of speech in patients with HD and argue that the studies cited by Drs. Rusz and Tykalova relied on acoustic measures that may only weakly correlate with perceptual ratings of dysarthric speech. They note that the trained raters were asked to rate speech relative to what they considered normal, and that their study identified CAG repeats and Unified HD Rating Scale motor scores as differing significantly across speech subgroups. This exchange highlights potential challenges with reconciling acoustic-based and rater-based phenotyping of speech patterns in neurologic disease.

Methodology Matters: The Impact of Research Design on Conversational Entrainment Outcomes.

Purpose Conversational entrainment describes the tendency for individuals to alter their communicative behaviors to more closely align with those of their conversation partner. This communication phenomenon has been widely studied, and thus, the methodologies used to examine it are diverse. Here, we summarize key differences in research design and present a test case to examine the effect of methodology on entrainment outcomes. Method Sixty neurotypical adults were randomly assigned to experimental groups formed by a 2 × 2 factorial combination of two independent variables: stimuli organization (blocked vs. random presentation) and stimuli modality (auditory-only vs. audiovisual stimuli). Individuals participated in a quasiconversational design in which the speech of a virtual interlocutor was manipulated to produce fast and slow speech rate conditions. Results There was a significant effect of stimuli organization on entrainment outcomes. Individuals in the blocked, but not the random, groups altered their speech rate to align with the speech rate of the virtual interlocutor. There were no effect of stimuli modality and no interaction between modality and organization on entrainment outcomes. Conclusion Findings highlight the importance of methodological decisions on entrainment outcomes. This underscores the need for more comprehensive research regarding entrainment methodology.

An Acoustic Characterization of Prosodic Differences in Autism Spectrum Disorder and First-Degree Relatives

This study examined prosody through characterization of acoustic properties of the speech of individuals with ASD and their parents, during narration. A subset of utterances were low-pass filtered and rated for differences in intonation, speech rate, and rhythm. Listener ratings were minimally related to acoustic measures, underscoring the complexity of atypical prosody in ASD. Acoustic analyses revealed greater utterance-final fundamental frequency excursion size and slower speech rate in the ASD group. Slower speech rate was also evident in the ASD parent group, particularly parents with the broad autism phenotype. Overlapping prosodic differences in ASD and ASD Parent groups suggest that prosodic differences may constitute an important phenotype contributing to ASD features and index genetic liability to ASD among first-degree relatives.

The realization of English rhythm by Busan Korean speakers*

The purpose of the current study is to investigate the realization of speech rhythm in English as spoken by Korean learners of English. The study particularly aims to examine the rhythm metrics of English read speech by learners who speak Busan or the South Kyungsang dialect of Korean. Twenty-four learners whose L1 is Busan Korean and eight native speakers of English read a passage wherein five sentences were segmented and labeled as vocalic and intervocalic intervals. Various rhythm metrics such as %V, Varcos, and Pairwise Variability Indexes (PVIs) were calculated. The results show that Korean learners read English sentences with significantly more vocalic and consonantal intervals at a slower speech rate than native English speakers. The analyses of rhythm metrics revealed that when the speech rate was not normalized, Korean learners’ English showed more variability in the length of consonantal and vocalic intervals. However, speech-rate-normalized rhythm metrics for vocalic intervals indicated that Korean learners transferred their L1 rhythmic structures (a syllable-timed language) into their L2 speech (a stress-timed language). Overall, the results suggest that Korean learners’ English reflects the rhythmic characteristics of their L1. The effect of the learners’ L1 dialect on the realization of L2 speech rhythm is also speculated.

Going With the Flow: An Examination of Entrainment in Typically Developing Children.

Purpose Conversational entrainment is the tendency for individuals to modify their behavior to more closely converge with the behavior of their communication partner and is an important aspect of successful interaction. Evidence of entrainment in adults is robust, yet research regarding its development in children is sparse. Here, we investigate the emergence of entrainment skills in typically developing children. Method Data were collected from a total of 50 typically developing children between the ages of 5 and 14 years. Children participated in a quasiconversational paradigm with a virtual interlocutor. Speech rate of the interlocutor was digitally manipulated to produce fast and slow speech rate conditions. Results Data from the fast and slow conditions were compared using linear mixed models. Results indicated that children, regardless of age, did not alter their speech to match the rate of the virtual interlocutor. Conclusions Findings suggest that entrainment in children may not be as robust as entrainment in adults and therefore not adequately captured with the current experimental paradigm. Modifications to the current paradigm will help identify a methodology sufficiently sensitive to capture the speech alignment phenomenon in children and provide much needed information regarding the typical stages of entrainment development.

Comparison of Speech Rate and Long-Term Average Speech Spectrum between Korean Clear Speech and Conversational Speech.

Background and ObjectivesClear speech is an effective communication strategy used in difficult listening situations that draws on techniques such as accurate articulation, a slow speech rate, and the inclusion of pauses. Although too slow speech and improperly amplified spectral information can deteriorate overall speech intelligibility, certain amplitude of increments of the mid-frequency bands (1 to 3 dB) and around 50% slower speech rates of clear speech, when compared to those in conversational speech, were reported as factors that can improve speech intelligibility positively. The purpose of this study was to identify whether amplitude increments of mid-frequency areas and slower speech rates were evident in Korean clear speech as they were in English clear speech.Subjects and MethodsTo compare the acoustic characteristics of the two methods of speech production, the voices of 60 participants were recorded during conversational speech and then again during clear speech using a standardized sentence material.ResultsThe speech rate and longterm average speech spectrum (LTASS) were analyzed and compared. Speech rates for clear speech were slower than those for conversational speech. Increased amplitudes in the mid-frequency bands were evident for the LTASS of clear speech.ConclusionsThe observed differences in the acoustic characteristics between the two types of speech production suggest that Korean clear speech can be an effective communication strategy to improve speech intelligibility.

Effects of utterance rate and length on the spatiotemporal index in Parkinson’s disease

Purpose: To characterise labial articulatory pattern variability using the spatiotemporal index (STI) in speakers with idiopathic Parkinson’s disease (PD) across different speaking rates and syllable-sentence conditions compared to age- and sex-matched healthy controls.Method: Ten speakers with mild–severe idiopathic PD and 10 controls produced “pa” and the Rainbow Passage at slow, typical and fast speech rates. Upper lip and lower lip kinematics were digitised during a motion capture system. Data were analysed using linear mixed modelling.Result: Regardless of the participant group, a high STI value was observed in the fast speech rate for the “pa” syllable condition, particularly for movements of the lower lip. As utterance rate increased, the control group showed the highest variability, followed by PD OFF and PD ON conditions. Syllable “pa” showed a greater STI value compared to both the first and second utterance of Rainbow Passage.Conclusion: PD manifests sufficient residual capacity to achieve near-normal motor compensation to preserve the consistency of lower lip movements during speech production. The lack of a significant difference in lip STI values between ON-OFF medication states suggests that dopaminergic treatment does not influence stability of speech for individuals with mild-moderate stage PD.

Six-second rule revisited

Abstract The most famous rule on the speed of subtitles is the six-second rule. In this study we investigate if the six-second rule is too slow for contemporary viewers. We also address the question of whether subtitle processing depends on the speech rate of film dialogues and on viewer’s proficiency in the language of the film soundtrack. With these questions in mind, we tested 53 Polish viewers watching English videos at two different speech rates (slow and fast), subtitled into Polish in accordance with the six-second rule. We examined participants’ reading patterns and comprehension and asked them to assess subtitle speed and the congruity of subtitles with the dialogue. Analysing people’s eye movements enabled us to measure that viewers were looking at subtitles for only about 30% of the subtitle display time. We found that the film speech rate affected comprehension: faster dialogues, implying more text condensation in subtitles, resulted in lower comprehension compared to slow speech rates. Viewers more proficient in the language of the film soundtrack spent less time gazing at subtitles than those who had only elementary knowledge of the language.

Correlations between linguistic change and linguistic performance among heritage speakers of Danish in Argentina

Abstract The article addresses the issue of an assumed correlation between heritage language speakers’ linguistic system and their fluency. Previous research has shown that heritage language speakers who have grammatical patterns that are divergent from the language spoken in the homeland also have lexical retrieval problems and speak with a slower speech rate. We approach the issue by examining two linguistic features and 11 different fluency features (‘structural’ and ‘performance features’) of 10 speakers of Argentine Heritage Danish. A factor analysis shows that the 11 performance features can be reduced to 2 factors that characterize the speakers ‘lexical retrieval’ and ‘fluency’. A correlation analysis with the two structural features, weakening of a phonological contrast between /oː/ and /ɔː/ and simplification of the gender system, shows that there are no correlations at all between these two measures of the speakers’ language production, in contrast to what previous research has found.

The effect of delayed auditory feedback (DAF) and frequency altered feedback (FAF) on speech production: cochlear implanted versus normal hearing individuals

ABSTRACTNormal auditory feedback contributes to moment-to-moment control of speech production. Effects of auditory feedback’s absence on hearing-impaired individuals are widely documented, but auditory perturbation has not been investigated. Our objective was to evaluate the effect of delayed auditory feedback (DAF) and frequency altered feedback (FAF) on speech production among prelingual cochlear implant (CI) users and normal hearing (NH) individuals, to evaluate CI users’ reliance on auditory feedback. Twenty young adults (10 CI, 10 NH), without developmental and cognitive impairments, participated in the study. Under variable auditory feedback conditions, speech production (spontaneous or reading aloud) was measured using speech rate, percentage of interruptions, fundamental frequency (F0), and relative intensity. Results showed that (1) both DAF and FAF caused slower speech rates and more interruptions while reading aloud, with DAF having larger effect; (2) altered feedback produced no differences between groups, except an increase in F0 for CI users during DAF; and (3) CI users’ ability to understand speech via phone and without lip-reading was positively correlated with performance under DAF. These findings suggest that auditory perturbation similarly affects speech production among prelingual CI users and NH individuals, indicating CI users depend on auditory feedback to the same degree as normal hearing individuals.

ТЕМПОРАЛЬНЫЕ ХАРАКТЕРИСТИКИ КИТАЙСКОЙ КОММЕРЧЕСКОЙ РАДИОРЕКЛАМЫ

This paper examines prosodic features of Chinese commercial radio advertising in its informationally important and unimportant segments. We also present a brief description of the perceptual study that allowed to identify the syllables that comprise the informational core of the advertisements. The comparative analysis was performed to determine the significance of a number of factors in the distribution of the information load in advertising. Among these factors were: 1) beginning / middle / end of the utterance, 2) fast speech rate / slow speech rate / stability of speech rate, 3) the average speech rate and duration of the advertisement, 4) pauses location before / after / at both sides / the beginning of the utterance / the end of the utterance, 5) presence of repeated elements in advertising, 6) male / female voice. The conclusion is made that the most important factors in the distribution of the information load are duration, position in the utterance, male voice and slowing down the speech rate.

Now listen to this! Evidence from a cross-spliced experimental design contrasting pressuring and supportive communications

Motivating communications are a frequent experience within daily life. Recently, it has been found that two types of motivations are spoken with distinct tones of voices: control (pressure) is spoken with a low pitched, loud tone of voice, fast speech rate, and harsh sounding voice quality; autonomy (support) is spoken with a higher pitched, quieter tone of voice and a slower speech rate. These two motivational tones of voice also differentially impact listeners' well-being. Yet, little is known about the brain mechanisms linked to motivational communications. Here, participants were asked to listen to semantically identical sentences spoken in controlling, neutral, or autonomy-supportive prosody. We also presented cross-spliced versions of these sentences for maximum control over information presentation across time. Findings showed listeners quickly detected whether a speaker was providing support, being pressuring, or not using motivating tones at all. Also, listeners who are pressured do not seem to respond anew when a supportive motivational context arises, but those who had been supported are affected by a newly pressuring environment. Findings are discussed in light of motivational and prosody literatures, and in terms of significance for the role of motivational communications on behavior.

Clinical Progression in Four Cases of Primary Progressive Apraxia of Speech.

This case series details the clinical progression of patients with primary progressive apraxia of speech (PPAOS) to illustrate, using several methods and supplemental material examples, the changes that occur in speech and language functioning in this patient population. Four patients who presented with PPAOS were followed between 5 and 6 years. Two patients had predominant articulatory abnormalities (termed phonetic PPAOS), 1 had predominant prosodic abnormalities (prosodic PPAOS), and 1 had relatively equal articulatory and prosodic abnormalities (mixed PPAOS). Detailed speech (including acoustics), language, neurologic, and neuropsychological data were collected. At initial exam, the patients ranged from 60 to 77 years old, with presenting disease duration of 1.5-10 years. Although all patients presented with an isolated apraxia of speech, all developed varying degrees of aphasia and dysarthria. Patients with phonetic PPAOS developed relatively more severe aphasia than the other 2 patients. All patients eventually had severe functional communication limitations and required alternative or augmentative means of communication, although at varying times postonset of their initial speech problem. Two patients developed dysphagia, 3 showed mild-moderate Parkinsonism, and 2 developed depression. For all patients, simple temporal acoustic measurements documented slowed speech rate over time. This case series demonstrates that patients who initially present with PPAOS may develop aphasia and dysarthria, cognitive and behavioral changes, and other neurologic signs. Whether these changes can be predicted by the perceptual characteristics of the apraxia of speech is yet to be determined. The detailed longitudinal profiles provide valuable clinical insight into the progression of disease in people with PPAOS. https://doi.org/10.23641/asha.7051616.

A first investigation of tongue, lip, and jaw movements in persons with dysarthria due to multiple sclerosis

BackgroundMultiple sclerosis can affect the speech motor system and result in dysarthria. ObjectivesThis pilot study sought to identify tongue, lip, and jaw motor deficits in persons with dysarthria due to multiple sclerosis (PwDMS) to better understand the speech motor mechanisms that underlie their aberrant speech. MethodsTongue and jaw movements during “ai” and lower lip and jaw movements during “bob”were examined in eleven PwDMS and fourteen age- and sex-matched controls using three-dimensional electromagnetic articulography. Movement duration, maximum displacement, peak speed, stiffness (i.e., peak speed/displacement ratio), and jaw contribution to lower lip and tongue displacements were of particular interest. ResultsWhereas most kinematic measures yielded significant between-group differences for tongue and jaw motor performance during “ai”, lower lip and jaw motor performance during “bob” were mostly comparable between groups. ConclusionFindings suggest that speech movements of the tongue are differentially more impaired than those of the lower lip in PwDMS. Particularly the ability to move the tongue with adequate speed during speech was significantly impaired in PwDMS, which may explain, in part, their slowed speech rate. Aberrant jaw kinematics during “ai” may be a compensatory strategy to maximize speech clarity in the presence of the impaired tongue motor performance.

Depression recognition based on speech analysis

Depression is one of the common mental diseases. Patients with depression often have depressed moods such as sadness, guilty, low self-esteem, loss of interest, hypofunction and so on. They suffer from serious emotional problems, unexplained suffering, which has caused enormous losses to individuals, families and society. According to the World Health Organization, there are aproximately 322 million people suffering from depression in the whole world in 2017. While there are about 54 million depressive patients in China. Depression can be cured effciently. However, due to the complexity of the pathogenesis of depression, clinical diagnosis is accompanied with many difficulties. Firstly, the mental disease, especially depression, are not getting enough attention and even being misinterpreted by other people. Secondly, the depression patients are less willing to ask for help. Thirdly, it is hard to select and dignose the potential depression patients precisely, as well as there are limited medical resource for depression diagnosis. It is necessary to find a more convenient, objective and efficient way to assist the fast identification of depression. As a relatively objective and easily accessible variable, speech has its potential value. The speech of patient is easy to acquire, and also, it has been proved that the sound of depressed patients have special charcteristics such as slow speech rate, lack of cadence and so on. The purpose of this paper is to explore the relationship between speech and depression by establishing classification models of voice feature and depression prediction. In this research, 3(emotion mood: positive, neutral, negative)×3(task type: question answering, text reading, picture description) experimental design was employed, and the voice data was collected from the speech of individuals recorded during different tasks. 103 participants were inculded in this study, including 45 depression patients (age: 23.8–44.6, M =34.2, SD =10.4, males=22, females=23) and 58 healthy ones (age: 20.1–41.7, M =30.9, SD =10.8, males=27, females=31). The former were recruited in the hospital in Beijing Anding Hospital and Huilongguan Hospital, while the latter were recruited by advertisement. All of them were diagnosed by specialist with DSM-IV and MINI interview. All participants did not have substance abuse, substance dependence, personality disorders and other mental diseases, no serious physical illness or suicidal behavior. The education level of subjects are all above the elementary school. 988 Voice features were extracted from the speech data using open SMILE software. Logistic regression, a machine learning method, was used to train the predicting models. Results showed that the precision rate of predicting can reach to 82.9%. Based on machine learning methods, this paper employed voice features to establish predicting models of depression. Results show the speech of depression patients has certain predicting effect, which paves the way for the further identification of depression in a more thorough way.

Speech rate adjustment of adults during conversation

PurposeSpeech rate convergence has been reported previously as a phenomenon in which one's speech rate is influenced by his/her partner's speech rate. This phenomenon has been demonstrated in artificial settings, and to some extent, in mother-child interactions. The purpose of this study was to explore speech rate adjustment in a quasi-natural adult-adult conversation. MethodsAn A-B-A-B paradigm was used, in which ten adults conversed on a given topic with two experimenters. Speech rates of both communication partners were measured. ResultsParticipants significantly reduced their speech rate, in response to the experimenters’ reduction in speech rate. However, the participants' reduction in speech rate was significantly smaller than the experimenters’ reduction in speech rate. In addition, during the controlled slow speech rate, the participants’ speech rate correlated negatively with that of the experimenters’. ConclusionResults suggest that speech rate convergence is a non-linear phenomenon, and may be affected by various linguistic as well as communicational factors. From a clinical perspective, the results support the use of the modeling strategy in speech therapy, as a means to facilitate a reduction in clients' speech rate.