Junfeng Wang
Shuyu Yang- College of Design and Innovation, Shenzhen Technology University, Shenzhen, China
Voice user interface (VUI) is widely used in developing intelligent products due to its low learning cost. However, most of such products do not consider the cognitive and language ability of elderly people, which leads to low interaction efficiency, poor user experience, and unfriendliness to them. Firstly, the paper analyzes the factors which influence the voice interaction behavior of elderly people: speech rate of elderly people, dialog task type, and feedback word count. And then, the voice interaction simulation experiment was designed based on the wizard of Oz testing method. Thirty subjects (M = 61.86 years old, SD = 7.16; 15 males and 15 females) were invited to interact with the prototype of a voice robot through three kinds of dialog tasks and six configurations of the feedback speech rate. Elderly people’s speech rates at which they speak to a person and to a voice robot, the feedback speech rates they expected for three dialog tasks were collected. The correlation between subjects’ speech rate and the expected feedback speech rate, the influence of dialog task type, and feedback word count on elderly people’s expected feedback speech rate were analyzed. The results show that elderly people speak to a voice robot with a lower speech rate than they speak to a person, and they expected the robot feedback speech rate to be lower than the rate they speak to the robot. There is a positive correlation between subjects’ speech rate and the expected speech rate, which implies that elderly people with faster speech rates expected a faster feedback speech rate. There is no significant difference between the elderly people’s expected speech rate for non-goal-oriented and goal-oriented dialog tasks. Meanwhile, a negative correlation between the feedback word count and the expected feedback speech rate is found. This study extends the knowledge boundaries of VUI design by investigating the influencing factors of voice interaction between elderly people and VUI. These results also provide practical implications for developing suitable VUI for elderly people, especially for regulating the feedback speech rate of VUI.
1. Introduction
Research in speech recognition began in the 1950s. Early technology only recognizes the English pronunciation of 10 numerals (Li, 2018; Shahrebaki et al., 2018). Current voice interaction systems have been employed in many devices that feature human-computer interaction technology. It can recognize complete human natural language utterances, forming a significant and booming market segment (Zen et al., 2013). In the process of voice interaction, the user activates the device by speaking specific voice commands. After receiving the commands, the device recognizes the commands and gives feedback. The feedback is then transformed into a sound that simulates a human voice with artificial speech synthesis technology and plays through the speaker to form voice feedback (Hess and Zellman, 2018; Singh et al., 2021).
The visual interaction interface necessitates learning its operation (Page, 2014), understanding the meaning of graphical elements, and searching for the object to be operated within the visual range, all of which lead to high learning costs (Liu and Ma, 2010; Bai et al., 2020). Voice interaction, in contrast, only requires users’ short-term memory and clear verbal expression, with low learning costs. Thus, it is appropriate for children, the elderly, and people with visual impairment (Huang and Liu, 2017; Zhang and Zhang, 2019; Pradhan et al., 2020; Hua, 2021; Guo et al., 2022). Some scholars have applied it to designing and developing products for the elderly. For instance, Antonio et al. (2014) built an intelligent system with multichannel interaction for the elderly by collecting a language database of them and constructing an elderly-specific language recognizer. Jia (2018) researched the elderly using the SJTU user research system and constructed corresponding voice interaction application scenarios for them. Furthermore, they concluded that when completing task-oriented dialogs, short and simple dialogs can lessen the memory load of the elderly and increase task completion, according to the experiment.
However, the design of current voice interaction systems has not considered the reduced hearing, cognitive and comprehension abilities of the elderly, let alone specific adjustments to the feedback speech rate of the system (Murad et al., 2019). This causes the issue that older users frequently have difficulty hearing or remembering while using voice interaction devices (Czaja et al., 2006; Lee and Coughlin, 2015), which significantly negatively impacts their interaction effectiveness and user experience (Baba et al., 2004; Lee, 2015; Zhang, 2021). This paper takes the feedback speech rate of voice interaction systems for the elderly as the research object. The feedback speech rate that the elderly expect in different task scenarios is collected through voice interaction simulation experiments. The speech rate regulation strategy of the voice interaction system is constructed accordingly to improve the efficiency and user experience of the elderly when using the system.
2. Literature review
2.1. Speech rate of the elderly
Communication is a process in which both the speaker and the addressee express information by language, and its purpose is to convey information. Speech accommodation theory (SCT) is a sociolinguistic theory developed recently (Yuan, 1992; Ma, 1998). The theory suggests that the addressee’s speech acts characteristics can be a reference standard for the speaker to regulate speech acts. Speech convergence is a phenomenon that in daily conversation, the speaker’s speech pattern (diction, speech rate, grammar, phonology, et al.) is influenced by the addressee’s speech pattern and adjusted to it to gain the addressee’s approval and affirmation (Beebe and Giles, 1984).
Speech convergence is more likely to occur when the speaker is a subordinate or junior, and the conversation is serious. In this situation, the speaker’s speech pattern will converge with those of the addres(see Brennan and Clark, 1996; Barr and Keysar, 2002; Gijssels et al., 2016). Kemper’s study found that when young people converse with the elderly, they will slow their speech rate and reduce the length and complexity of the discourse to help the elderly understand the message, i.e., speech regulation mechanisms toward the elderly (Kemper, 1994). A study by Jiang (2017) noted that young people adopt speech convergence in telephone conversations with aged people, including but not limited to speech rate, average sentence length, and pause duration. The aim is to assist aged people in understanding the message, gain their trust, and enhance emotional intimacy.
Human-computer voice interaction design aims to simulate human-to-human verbal communication (Murad et al., 2019). Many intelligent voice devices define the voice interaction system as the user’s “intelligent voice assistant” (Rakotomalala et al., 2021) and refer to the user as the “master” during the conversation. This design reflects the relationship of ownership, subordination, and domination between the user and the voice device, comparable to the relationship between superiors and subordinates in interpersonal interactions (Nass et al., 1994; Powers and Kiesler, 2006; Ostrowski et al., 2022). On the other hand, voice interaction devices are inanimate objects, and users have lower intimacy and trust during initial use. They also interact more cautiously, which is not conducive to conveying information effectively (Dautenhahn, 2004; Höflich and El Bayed, 2015; Song et al., 2022). Thus, it is reasonable and essential to apply speech convergence strategies to improve the dialog relationship between voice interaction devices and users.
Based on speech accommodation theory, when elderly people interact with the system through voice, the system adjusts speech acts characteristics of the feedback following the elderly’s, which will enhance the emotional experience of the elderly (Myers et al., 2018, 2019). The paper analyzes the correlation between the speech rate of the elderly and their expected feedback speech rate. It also summarizes the feedback speech rate regulation strategies of the voice interaction system toward elderly people based on experimental results.
2.2. Dialog task type
In existing research, the types of dialog tasks were divided into non-task-oriented dialogs and task-oriented dialogs according to the different purposes (Luna-Garcia et al., 2018) for which users use voice interaction devices (Chen et al., 2017).
Non-task-oriented dialogs primarily refer to forms of interaction in which users have no clear expectations or specific goals about the feedback from the voice interaction system. Typical applications include listening to music, stories, and operas. The device recognizes the type of content users want to listen to after activating the device with the wake-up word. Then the device starts to play the audio, and users enter the listening stage. Users do not initiate the voice interaction process again until the audio finishes or the audio is dissatisfying. In this usage scenario, the user’s purpose is to pass the time and relieve loneliness. When interacting with devices, users can identify only some of the information intentionally or memorize it. Users only must confirm that the device is playing the needed content, which requires a low cognitive load.
Task-oriented dialogs mainly refer to voice interaction systems that assist users in accomplishing specific tasks, such as checking the weather, making hotel or restaurant reservations, et al., by single or multiple rounds of dialogs. With a specific goal, users activate the device and give the corresponding voice command when utilizing such functions. Then users extract the information needed from the feedback played by the device and store it in short-term or long-term memory, which can be applied to the specific task. A more complex situation is that the voice interaction system conveys information to users through multi-round dialogs, such as multi-round quizzes and quiz games. In these dialogs, users must mobilize brain functions such as thinking and memory to participate in the interaction. They must also comprehend and analyze the received feedback to respond with the highest cognitive load (Sayago et al., 2019; Moore and Urakami, 2022).
As shown in Table 1, this paper conducts a voice interaction simulation experiment based on the two kinds of dialog tasks to analyze the influence of dialog task types on the expected feedback speech rate for the elderly. However, people interact with VUI with no clear goals, they also execute a dialog task. It is a little bit confusing to name this kind of interaction as a non-task-oriented dialog. So, goal-oriented, and non-goal oriented are used to describe the interactions that with and without a clear goal, respectively.
Table 1. Functions and dialog tasks classifications of voice interaction devices.
2.3. Word count of single feedback
Human brain nerve cells gradually diminish after the age of 50 (Svennerholm et al., 1997; Scahill et al., 2003). Although the brain’s basic functions can generally be maintained, the volume of brain tissue may atrophy to varying degrees in some individuals, resulting in memory loss and personality changes. Brain nerve cells also impact the regulatory role of the brain’s involvement in other organs, affecting other organs’ functional performance (Wang, 2002). Compared to young people, the perceptual abilities of the elderly become blunted, and degenerative changes may occur in vision, visual perception, hearing, and auditory perception (Hawthorn, 2000; Yuan et al., 2006; Wilkinson and Cornish, 2018).
Elderly people find it more challenging to accept new things (Ziman and Walsh, 2018; Kowalski et al., 2020) and technologies (Kalimullah and Sushmitha, 2017) and acquire external information due to the deterioration in their perceptual abilities. Therefore, voice interaction simulation experiments were designed based on the feedback of different word counts from the voice interaction system to investigate the impact of the word count in single feedback on the elderly’s expected feedback speech rate.
2.4. Voice user interface for elderly people
Voice activated human machine interaction has developed rapidly in recent years. This has attracted extensive attention from the academic community. VUIs offer elderly people multiple advantages over traditional GUI/hardware interfaces by being fewer motor skills required, efficient of getting information, intuitive to interact, and rich of meaning through tone, volume, intonation, and speed (Sayago et al., 2019). More researchers are involved in the research of VUI for elderly people.
Ziman and Walsh (2018) studied the factors affecting seniors’ perceptions of VUI, and indicated that familiarity, usability, habit, aversion to typing, and efficiency of voice input are the most critical factors that influenced the seniors’ perceptions and acceptance of VUI. Research (Pradhan et al., 2020; Song et al., 2022) focused on the adoption and usage of VUI by elderly people with low technology use found that perceived usefulness, perceived ease of use, and trust are decisive factors that determined elderly people adapt VUI or not and influenced their attitudes to VUI. After studied the patterns of tactics that people employ to overcome the problems when interacting with VUI, Myers (Myers et al., 2018) indicated that feedback strategy could be a worthy point to improve VUI’s user experience.
Meena et al. (2014) proposed a data-driven approach to building models for online detection of suitable feedback response locations in the user’s speech. The results from the user evaluation through human computer interaction show that the model trained on speaker behavioral cues offers both smoother turn-transitions and more responsive system behavior. Other research that concentrated on feedback position during conversation identified feedback locations through multimodal models (Boudin et al., 2021), Interdisciplinary Corpus (Boudin, 2022), and voice activity (Truong et al., 2010; Ekstedt and Skantze, 2022).
3. Methods
This study explores the influence of dialog type and feedback word count on the user’s expected feedback speech rate.
3.1. Research design
This study applies Wizard of Oz testing to conduct voice interaction simulation experiments (Cordasco et al., 2014; Ma et al., 2022). Wizard of Oz testing is a method in which the tester acts as a “wizard” to manipulate the object to be tested to make it interact with the subject and collects relevant experimental data (Dahlbäck et al., 1993). This method is widely applied to study the usability and user acceptance of voice interaction systems, natural language applications, command languages, imaging systems, and pervasive computing applications in the prototype stage (Shin et al., 2019). In the development of voice-interactive product design, Wizard of Oz can assist UX researchers in cutting costs by quickly testing the usability of products at different stages (White and Lutter, 2003). Some studies used the Wizard of Oz testing to investigate the elderly’s acceptance of smart home products equipped with voice-interactive systems (Portet et al., 2013; Porcheron et al., 2020).
Before the experiment, dialog tasks between users and voice interaction devices were designed based on the current mainstream voice interaction process. During the experiment, subjects are first informed about the experiment’s background and the task requirements. Then subjects perform the appropriate voice interaction behavior in line with the task requirements, and the experimenter manipulates the product to provide feedback. The assistant collected subjects’ speech rate data during the process and the subjects’ expected feedback speech rate with the speech rate satisfaction scale (Ghosh et al., 2018). The assistant also needs to observe and record subjects’ performance and attitudes. Finally, user interviews and quantitative analysis were conducted to explore the users’ expectations of the feedback speech rate under different task scenarios (Iniguez et al., 2021).
3.2. Subjects and settings
To select the qualified subjects, the experimenter explained the process and conducted a pre-talk test to ensure that the subjects could hear the conversation clearly in the experimental environment and understand the overall process before the experiment. Thirty elderly people with normal hearing and no obvious cognitive impairment were chosen for the study; 10 were between 60 and 70 years old, six were 70 years old or above, and the others were 50 and 60.
The experiment was conducted in a quiet indoor environment with soft light (Figure 1). The experimenter had previously explained the experimental task process to the subjects before the experiment. The subjects sat facing the 15.6-inch display, a Philips SPA510 dual-channel speaker was used to play the voice interaction content, and a Newman ZM02 microphone collected the user’s voice commands.
Figure 1. Environment for voice interaction simulation.
3.3. Materials
3.3.1. Dialog tasks
Currently, intelligent voice interaction devices are widely used in daily life. The cognitive load of the elderly is low when engaging in non-goal-oriented conversations while high when engaging in goal-oriented conversations.
To investigate the correlation between the user’s expected feedback speech rate and the type of dialog task, Task 1 is set as a non-goal-oriented dialog in the scenario that the elderly was bored at home and wanted to listen to the radio for entertainment. It is a light-load interaction task. Task 2 is a goal-oriented dialog, with the scenario of elderly people checking the weather the next day before going out, which was a heavy-load interaction task. In order to find the relationship between the user’s expected feedback speech rate and the word count in single feedback, Task 3 is also designed as a goal-oriented dialog in which the scenario is that the elderly is checking the next day’s schedule. The word count in the feedback of Task 3 is different from that in Task 2.
In a text-to-speech system, the length of punctuation that pauses in the corpus is approximately equal to a word. So, the punctuation in the corpus is counted into the total words. Table 2 shows the three dialog tasks. The voice dialog is in Chinese, so the feedback words of the three tasks are also counted in Chinese characters.
Table 2. Dialog task of voice interaction experiment.
To avoid sequential effects, a set of experiments was conducted with dialogs tasks in three scenarios, including Task 1, “listen to the news,” Task 2, “check the weather,” and Task 3, “check the schedule.” Each subject was required to complete six experiments with six random gears of feedback speed for all tasks.
3.3.2. Feedback speech rate
Speech rate typically refers to the speed of articulation, while it can also refer to the auditory perceptual impression of the pacing of words (Cao, 2003). The acceptable speech rate is below 300 words/min, and over this range, listeners may have difficulty following the conversation (Portet et al., 2013). Besides, a speech rate of 100–150 syllables/min is considered a “super-slow speech rate,” which is rare in daily conversations (Meng, 2006). Therefore, six speech rates were defined as the feedback speed configurations for the experiment, as shown in Table 3. The six feedback speech rates of the system were 2.25 words/s (135 words/min), 2.75 words/s (165 words/min), 3.25 words/s (195 words/min), 3.75 words/s (225 words/min), 4.25 words/s (255 words/min) and 4.75 words/s (285 words/min).
Table 3. Configurations of feedback speech rate.
3.3.3. Feedback corpus
The text of the feedback corpus was synthesized into speech using the Swift text-to-speech system, and the pronunciation source was the built-in standard female voice with a 16 khz sampling rate. Figure 2 depicts the process of the feedback corpus synthesizing. Firstly, the text of the feedback corpus was designed according to the purpose of the dialog. Secondly, each speech was generated by Swift, the text-to-speech system. Thirdly the synthesized speech was imported into Adobe Audition to modify the speech rate according to the configurations in Table 3, and then export the corpus for the experiment.
Figure 2. The process of synthesizing the feedback corpus.
3.4. Experimental process
The experiment was designed based on Wizard of Oz testing, in which the subject initiated the voice interaction, and the experimenter acted as a “wizard” to operate the prototype of the voice interaction system to give feedback. The prototype was made of an avatar and synthesized feedback corpus. To guide subjects, we made some slides incorporated with the prototype (see Figure 3). The dialog between subjects and voice robot prototype was simulated by switching the slides.
Figure 3. The Prototype of the voice interaction system.
Take Task 1, “listen to the news,” as an example. The experimenter explained the experimental process to the subject before it started. To ensure the experiment ran smoothly, the subjects were informed of the simulated scenario of the dialog task and activated the voice interaction system in the pre-test. Slide (1) is the experiment guide, showing the subjects the dialog scenario and wake-up words. After issuing the wake-up word, subjects entered slide (2), and the experimenter played feedback, “Hey, I’m here!” indicating that the system was activated. The slide (3), an experimental guide, was shown to suggest to the subjects the dialog scenarios and voice commands to be issued. During the experiment, to enhance the subjects’ immersion, waiting and speaking animations were added to the icons of the robot.
The experiment was conducted with a 5-point Likert scale to evaluate subjects’ satisfaction with the feedback speech rate of the system (Feng, 2002). Subjects’ satisfaction was calculated based on the options corresponding to the ratings in Table 4. A higher satisfaction score of the feedback speech rate indicates a higher acceptance by the user.
Table 4. Speech rate satisfaction scale.
While performing the experimental task, the subjects’ speech content was recorded with Adobe Audition CC2020 and a microphone in 16 kHz, mono, 32-bit audio format.
When subjects finished all dialog task, some open-ended questions were asked. The mainly purpose of this post-experiment interview is to know the attitudes and using problems of elderly people about VUI and voice robot. The questions include: “what do you think about the experience of talking with VUI or voice robot?,” “Did you have any problems when you talking with VUI or voice robot?,” “Did you always understand what the VUI or voice robot said? If not, what makes it incomprehensible?”
4. Results
4.1. Speech rate
The number of words per second during the subjects’ speech was defined as the subject’s speech rate, noted as Vs in words/s (Li et al., 2019). Subjects’ raw recordings were imported into Adobe Audition 2020 for further processing. According to the speech rate test methodology proposed by Kim et al. (2015), subjects’ single complete speech was intercepted from the recording for speech rate analysis. If the subject paused for longer than 2 s in a single speech, the pause was deleted to obtain the corpus in the stable state, which was used to extract the subject’s speech rate. If the pause time was between 1 and 2 s, it was necessary to determine whether the subjects showed obvious doubt or nervousness according to the recorded video and choose the corpus that the subject was in a stable situation. If the subject paused for less than 1 s, it was considered normal.
The number of words is denoted as S and the total duration after deleting the silent pauses is denoted as T in single speech. The subjects’ speech rate Vs during the voice interaction is calculated by Equation (1).
Before the experiment started, speeches subjects spoke to the experimenter were recorded. Six sentences were extracted to analyze the speech rates at which subjects speak to the experimenter. At the end of the experiment, six sets of speech rate data were collected from each subject separately. The mathematical expectation speech rates at which subjects speak to a person and a voice robot are shown in Figure 4. One subject did not complete the experiment, and then there were 29 subjects’ data. A paired t-test was performed on the two kinds of speech rates, and the results show a significant difference (p = 0.000), meaning subjects speak to a voice robot at a very different speech rate than a person.
Figure 4. Distribution of subjects’ speech rate.
4.2. Correlation between subjects’ speech rate and expected feedback speech rate
Experiments were conducted randomly with subjects on the three dialog tasks with six speech rates for feedback from the voice robot. After each session, subjects were asked to score the feedback speech rate on a satisfaction scale shown in Table 4. At the end of the experiment, the speech rate with the highest score was determined as the subjects expected feedback speech rate for this task. If more than one feedback speech rate configuration that got the highest score, then the mean of these configurations is defined as the expected feedback speech rate.
The subjects’ expected feedback speech rates of the three dialog tasks are, respectively, noted as W1, W2, and W3, which are shown in Figure 5. Paired t-test was performed on the speech rate at which subjects speak to a voice robot and the expected feedback speech rate. The results showed (Table 5) that there were significant differences (p = 0.00) between the two variables for all three dialog tasks. That means in all cases of dialog task type and feedback word count, subjects expected feedback speech rate lower than their speech rate at which they speak to a voice robot.
Figure 5. Distribution of subjects’ expected speech rate of feedback.
Table 5. Paired t-test of the subjects’ speech rate and the expected feedback speech rate.
Pearson correlation analysis was conducted on the subjects’ speech rate Vs and the subject’s expected system feedback speech rate W1, W2, W3, and the results are shown in Table 6. For Task 1, the results indicate no significant correlation (r = 0.113, p = 0.56) between the subjects’ speech rate and the expected feedback speech rate. For Task 2 and Task 3, there are significant positive correlations (r = 0.417, p = 0.025 and r = 0.399, p = 0.032) between the subjects’ speech rate and the expected feedback speech rate.
Table 6. Correlation between the subjects’ speech rate and the expected feedback speech rate.
The variations of the expected system feedback speech rate W2 and W3 with the subjects’ speech rate Vs are shown in Figures 6, 7 separately. A linear regression analysis was conducted to analyze the specific performance of the correlation between subjects’ speech rates and the expected feedback speech rates. The linear regression models are also shown in Figures 6, 7. The F-test (F = 5.670, p = 0.025; F = 5.122, p = 0.032) and T-test (t = 2.381, p = 0.025; t = 2.263, p = 0.032) also show a significance of the linear relationship of the regression model and the regression coefficient.
Figure 6. The variation of the expected feedback speech rate W2 with the subjects’ speech rate Vs.
Figure 7. The variation of the expected feedback speech rate W3 with the subjects’ speech rate Vs.
4.3. The influence of dialog task on expected feedback speech rate
LSD-t-test was conducted on the subjects’ expected feedback speech rates W1, W2, and W3. The results are shown in Table 7. The significance of the chi-square test was p = 0.866, which validates the homogeneity of the collected data. As shown in Table 8, a one-way ANOVA was carried out on the subjects’ expected feedback speech rates in different types of dialog tasks. The results indicate no significant difference (p = 0.065) among the three expected feedback speech rates across dialog task types. That means dialog task type did not significantly affect the expected feedback speech rate.
Table 7. Homogeneity test of variance for W1, W2, and W3.
Table 8. One-way ANOVA test for expected feedback speech rate in different types of dialog tasks.
The post hoc LSD-t-test was conducted on the users’ expected system feedback speech rate W1, W2, and W3, and the correlation between the data was examined in two pairs. The results are depicted in Table 9. There is no significant difference between W1 and W2 (p = 0.831) under the conditions of different types of dialog tasks and the same number of words of feedback, which also indicates that the dialog task type does not affect the elderly’s satisfaction with the expected feedback speech rate.
Table 9. Post hoc LSD-t test for W1, W2, and W3.
4.4. The influence of the feedback word count on expected feedback speech rate
The feedback word count of Task 1 and Task 2 was 37, and the mean value of the subjects’ expected feedback rate was 3.61 words/s and 3.63 words/s, respectively. The word count of Task 3 was 18, and the mean value of the expected feedback speech rate was 3.81 words/s.
As shown in Table 9, the subjects’ expected feedback speech rates W1 andW3 are significantly different (p = 0.024). W2 and W3 differ significantly (p = 0.039) alike. One-way ANOVA was also carried out on the subjects’ expected feedback speech rates in different word counts of dialog tasks. As shown in Table 10, the results indicate a significant difference (p = 0.005) among the three expected feedback speech rates across feedback word count. That means feedback word count affects the expected feedback speech rate significantly.
Table 10. One-way ANOVA test for expected feedback speech rate in different word counts of dialog task.
It suggests that there was a difference in the expected speech rate of feedback between Task 1, “listen to the news,” and Task 3, “check the schedule.” Under conditions of different types of tasks and word counts of feedback, there is a difference in subjects’ expected feedback speech rates. Thus, it is essential to determine the factors affecting feedback speech rate. It indicates a difference between the subjects’ expected system feedback speech rate when completing the dialog task of acquiring information in Task 2 and Task 3, i.e., task scenarios with the same task type but different feedback word counts.
It demonstrates that the amount of system feedback words impacts users’ assessments of the system feedback speech rate. Subjects preferred a slower system feedback speech rate in the task scenario with more words of feedback than the feedback with fewer words.
Data analysis and post-experiment user interviews found that when the system feedback contained more words, the elderly’s expected feedback speech rate was significantly slower than the scenario with fewer words of feedback. It indicates that elderly people have limited cognitive ability when processing feedback information. When receiving more feedback, they take more time to remember and store the information; therefore, elderly people expect the feedback speech rate to be slower.
5. Discussion
There is a significant difference between the speech rate subjects speak to people and the voice robot. When elderly people speak to a robot, they consciously slow their speech rate. Most elderly people are not very familiar with voice interaction technology and the product, which led to a slight on voice robot (Branigan et al., 2011; Koulouri et al., 2016). “I think the robot may not be as smart as people, so I speak to it with a lower speech rate to ensure it hears me clearly and understand me,” subject No. 14 said in the interview after the experiment. Whether this phenomenon exists in people familiar with voice user interfaces and artificial intelligence technology is pending.
Elderly people expect the voice robot to give feedback at a slower speech rate than their own. From the aspect of language expression, speech rate reflects one’s cognitive, understanding and memory skills (Rönnberg et al., 1989; Sanju et al., 2019; Lotfi et al., 2020; Huiyang and Min, 2022). Elderly people want their interlocutor to talk to themselves with a lower speech rate to ensure they can hear and understand the speaker clearly. Even the speaker is a robot. Meanwhile, elderly people with faster speech rates expect a faster feedback speech rate, which confirms the current study results that people with faster speech rates expect their interlocutor to respond with a faster speech rate (Brown et al., 1985; Hargrave et al., 1994; Freud et al., 2018). These suggest to some extent that the speech convergence is applicable to the interaction between elderly people and VUI.
Compared with non-goal-oriented dialog, goal-oriented dialog features specific information acquisition, which may require the listener to concentrate more on the speaker’s feedback (Galley, 2007; Zhang et al., 2018; Stigall et al., 2020). Nevertheless, the results show that dialog task type did not significantly affect the expected feedback speech rate. Subject No. 7 said, “regardless of whether I have a clear goal of information acquisition, I always hope to hear the voice clearly and try my best to understand what I heard.” Based on our observations of the subjects during the experiments, they always try their best to listen and remember the voice robot’s feedback, even though they are not required to do so. This may be slightly different from the scenario of the elderly listening to the radio and music in their leisure time. Nearly 83% of the subjects (N = 25) said that they did not and would not remember all details of the music and radio they listened to in their leisure time.
As mentioned above, subjects always try their best to remember the information of the feedback from the voice robot. It is reasonable that feedback word count significantly affects the expected feedback speech rate. Although the experimental scenario is different from the real scenario of voice user interface usage, we still believe that the voice user interface designers should set a reasonable feedback speech rate to ensure users can accurately capture all the content of the feedback.
6. Conclusion
This paper focuses on the effects of the elderly’s speech rate, types of voice interaction task, and the word count of feedback on elderly people’s expected feedback speech rate. It is found that the elderly’s speech rate and the word count in a single feedback have a significant influence on elderly people’s expected feedback speech rate. However, dialog task type affects the expected feedback speech rate inapparently. The faster elderly people speak, the faster feedback speech rate they desire, but not faster than their own. The more words of feedback are, the slower the elderly’s expected feedback speech rate is. These results also provide valuable implications for VUI user experience design. The feedback speech rate should be defined according to the interacting speech rate of elderly people and the word count of feedback content.
This study was designed for theoretical and practical application, especially the linear regression models of subjects’ speech rate and their expected feedback speech rate could be applied to developing a voice robot or other applications with the voice user interface. Besides, the word count of the feedback is another factor that should be considered when defining the feedback speech rate. In this study, two typical scenarios, which contain 18 and 37 Chinese words, are used in the experiment, respectively. The results show a significant difference in the expected feedback speech rates. However, these two numbers are not guidelines to follow. More research should focus on the effect of word count or information blocks on the expected feedback speech rate.
This study is carried out with Chinese people, and the materials are also made of Chinese characters and mandarin, so the results just explain the interaction between Chinese elderly people and voice robots. As different languages are spoken with different speech rates intrinsically, the problems this paper focuses on could be studied further in other languages.
Data availability statement
The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.
Ethics statement
The studies involving human participants were reviewed and approved by Ethics Committee of Shenzhen Technology University. The patients/participants provided their written informed consent to participate in this study.
Author contributions
JW: conceptualization, methodology, investigation, supervision, project administration, and writing – review and editing. SY and ZX: data curation. SY and JW: formal analysis. SY: visualization. ZX: materials. JW and SY: writing – original draft preparation. All authors have read and agreed to the published version of the manuscript.
Funding
This research was supported by the Humanities and Social Science Projects of the Ministry of Education of China (Grant No. 21YJC760078) and the Postgraduate Education and Teaching Reform Project of Guangdong Province of China (Grant No. 2022JGXM094).
Conflict of interest
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
Antonio, T, Annika, H, Jairo, A, Nuno, A, and Geza, N. (2014). Speech-centric multimodal interaction for easy-to-access online services – a personal life assistant for the elderly, In Proceeding of 5th International conference on software development and Technologies for Enhancing Accessibility and Fighting Info-Exclusion, 27, 389–397
Baba, A., Yoshizawa, S., Yamada, M., Lee, A., and Shikano, K. (2004). “Acoustic models of the elderly for large-vocabulary continuous speech recognition” in Electronics and Communications in Japan, Part II, vol. 87, Tokyo, 49–57.
Bai, X., Yu, J., Qin, L., and Yang, H. (2020). Cognitive aging of the elderly population and interaction Interface Design of Elderly Products, packaging. Engineering, 10, 7–12. doi: 10.19554/j.cnki.1001-3563.2020.10.002
Barr, D. J., and Keysar, B. (2002). Anchoring comprehension in linguistic precedents. J. Mem. Lang. 46, 391–418. doi: 10.1006/jmla.2001.2815
Beebe, L., and Giles, H. (1984). Speech-accommodation theories: a discussion in terms of second-language acquisition. Int. J. Sociol. Lang. 1984, 5–32. doi: 10.1515/ijsl.1984.46.5
Boudin, Auriane. (2022). “Interdisciplinary corpus-based approach for exploring multimodal conversational feedback,” in International conference on multimodal interaction. Bengaluru: ACM, 705–710.
Boudin, Auriane, Bertrand, Roxane, Rauzy, Stéphane, Ochs, Magalie, and Blache, Philippe. (2021). A multimodal model for predicting conversational feedbacks.” In Text, speech, and dialogue, (Eds.) Kamil Ekštein, František Pártl, and Miloslav Konopík, 12848:537–49. Cham: Springer International Publishing
Branigan, H. P., Pickering, M. J., Pearson, J., McLean, J. F., and Brown, A. (2011). The role of beliefs in lexical alignment: evidence from dialogs with humans and computers. Cognition 121, 41–57. doi: 10.1016/j.cognition.2011.05.011
Brennan, S. E., and Clark, H. H. (1996). Conceptual pacts and lexical choice in conversation. J. Exp. Psychol. Learn. Mem. Cogn. 22, 1482–1493.
Brown, B. L., Giles, H., and Thakerar, J. N. (1985). Speaker evaluations as a function of speech rate, accent and context. Lang. Commun. 5, 207–220. doi: 10.1016/0271-5309(85)90011-4
Cao, J. (2003). The characteristics and changes of speeches, Phonetic Research Report of Institute of Linguistics, Chinese Academy of Social Sciences, 6th Chinese Academic Conference on Modern Phonetics. Beijing, 143–148.
Chen, H., Liu, X., Yin, D., and Tang, J. (2017). A survey on dialogue systems: recent advances and new Frontiers. ACM SIGKDD Explor. Newsl. 19, 25–35. doi: 10.1145/3166054.3166058
Cordasco, G, Esposito, M, Masucci, F, Riviello, M, and Esposito, A, (2014). Assessing voice user interfaces: the vAssist system prototype In Proceeding 5th IEEE conference on cognitive infocommunications (coginfocom), New York, 91–96, 2014.
Czaja, S. J., Charness, N., Fisk, A. D., Hertzog, C., Nair, S. N., Rogers, W. A., et al. (2006). Factors predicting the use of technology: findings from the center for research and education on aging and technology enhancement (CREATE). Psychol. Aging 21, 333–352. doi: 10.1037/0882-7974.21.2.333
Dahlbäck, N., Jönsson, A., and Ahrenberg, L. (1993). Wizard of Oz studies — why and how. Knowl. Based Syst. 6, 258–266. doi: 10.1016/0950-7051(93)90017-N
Dautenhahn, K, (2004) Robots we like to live with⁈ – a developmental perspective on a personalized, life-long robot companion, RO-MAN 2004. In proceeding of 13th IEEE International workshop on robot and human interactive communication (IEEE catalog no. 04TH8759), 17–22
Ekstedt, Erik, and Skantze, Gabriel. (2022). Voice activity projection: self-supervised learning of turn-taking events. Interspeech 2022, 5190–5194 doi: 10.21437/Interspeech.2022-10955
Feng, X. (2002). Design of Questionnaire in social survey, Tianjin: Tianjin People’s Publishing House.
Freud, D., Ezrati-Vinacour, R., and Amir, O. (2018). Speech rate adjustment of adults during conversation. J. Fluen. Disord. 57, 1–10. doi: 10.1016/j.jfludis.2018.06.002
Ghosh, D, Foong, P, Zhang, S, and Zhao, S. (2018). “Assessing the utility of the system usability scale for evaluating voice-based user interfaces, the sixth international symposium of Chinese CHI,” in Association for Computing Machinery. New York, NY, USA, 11–15.
Gijssels, T., Casasanto, L. S., Jasmin, K., Hagoort, P., and Casasanto, D. (2016). Speech accommodation without priming: the case of pitch. Discourse Process. 53, 233–251. doi: 10.1080/0163853X.2015.1023965
Guo, H., Wang, H., and Yin, S. (2022). Adaptive aging design of smart TV VUI based on language cognition. Packag. Eng. 43, 50–54. doi: 10.19554/j.cnki.1001-3563.2022.08.007
Hargrave, S., Kalinowski, J., Stuart, A., Armson, J., and Jones, K. (1994). Effect of frequency-altered feedback on stuttering frequency at normal and fast speech rates. J. Speech Lang. Hear. Res. 37, 1313–1319. doi: 10.1044/jshr.3706.1313
Hawthorn, D. (2000). Possible implications of aging for interface designers. Interact. Comput. 12, 507–528. doi: 10.1016/S0953-5438(99)00021-1
Hess, A. S., and Zellman, J. (2018). Could you please repeat that? Speech design best practices for minimizing errors. Proc. Hum. Fact. Ergon. Soc. Annu. Meet 62, 1002–1006. doi: 10.1177/1541931218621231
Höflich, JR, and El Bayed, A. (2015). Perception, acceptance, and the social construction of robots—Exploratory studies, Social Robots from a Human Perspective, Cham: Springer International Publishing, 39–51.
Hua, J. (2021). Research on information interaction of digital reading products for preschool children based on user experience. Design 34, 121–123.
Huang, L., and Liu, T. (2017). Interactive innovation design of barrier-free products for the blind. Packag. Eng. 38, 108–113. doi: 10.19554/j.cnki.1001-3563.2017.24.023
Huiyang, S., and Min, W. (2022). Improving interaction experience through lexical convergence: the prosocial effect of lexical alignment in human-human and human-computer interactions. Int. J. Hum. Comput. Interact. 38, 28–41. doi: 10.1080/10447318.2021.1921367
Iniguez, A. L., Gaytan, L. S., Garcia-Ruiz, M. A., and Maciel, R. (2021). Usability questionnaires to evaluate voice user interfaces. IEEE Lat. Am. Trans. 19, 1468–1477. doi: 10.1109/TLA.2021.9468439
Jia, G. (2018). Research of AI speaker voice interaction design for the aged. Shenzhen: South China University of Technology.
Jiang, F. (2017). Speech regulation in telephone conversation between young and old people. Education 28:291.
Kalimullah, K, and Sushmitha, D. (2017). Influence of design elements in Mobile applications on user experience of elderly people, 8th International conference on emerging ubiquitous systems and pervasive networks (EUSPN 2017)/7th International conference on current and future trends of information and communication Technologies in Healthcare (icth-2017) / affiliated workshops, Amsterdam: Elsevier Science Bv, 113, 352–359
Kemper, S. (1994). Elder speak: speech accommodations to older adults. Neuropsychol. Cogn. Aging 1, 17–28. doi: 10.1080/09289919408251447
Kim, H., Wang, M., Zhang, Y., and Zhao, F. (2015). A study of speech speed in normal adults reading aloud and spontaneous speech. J. Audiol. Speech Disord 23, 240–243.
Koulouri, T., Lauria, S., and Macredie, R. D. (2016). Do (and say) as I say: linguistic adaptation in human–computer dialogs. Hum. Comput. Interact. 31, 59–95. doi: 10.1080/07370024.2014.934180
Kowalski, J, Jaskulska, A, Skorupska, K, Abramczuk, K, Kopec, W, and Marasek, K. (2020). Older adults and voice interaction: a pilot study with Google home. In Proceedings of the 31st Australian conference on human-computer-interaction (ozchi'19), New York: Assoc Computing Machinery, 423–427.
Lee, J. (2015). Aging and speech understanding. J. Audiol. Otol. 19, 7–13. doi: 10.7874/jao.2015.19.1.7
Lee, C., and Coughlin, J. (2015). PERSPECTIVE: older Adults' adoption of technology: an integrated approach to identifying determinants and barriers. J. Prod. Innov. Manage. 32, 747–759. doi: 10.1111/jpim.12176
Li, X. (2018). Overview of speech recognition technology based on human-computer interaction. Electron. World 21:105. doi: 10.19353/j.cnki.dzsj.2018.21.060
Li, Y., Wang, J., and Wang, W. (2019). Optimization of VUI feedback mechanism based on the time perception. Decor. Furnish. 7, 100–103. doi: 10.16272/j.cnki.cn11-1392/j.2019.07.023
Liu, H, and Ma, F. (2010). Research on visual elements of web UI design, IEEE 11th International conference on computer-aided Industrial Design & Conceptual Design 1, 428–430
Lotfi, Y., Samadi-Qaleh-Juqy, Z., Moosavi, A., Sadjedi, H., and Bakhshi, E. (2020). The effects of spatial auditory training on speech perception in noise in the elderly. Crescent J. Med. Biol. Sci. 7, 40–46.
Luna-Garcia, H., Mendoza-Gonzalez, R., Gamboa-Rosales, H., Celaya-Padilla, J., Galvan-Tejada, C., Lopez-Monteagudo, F., et al. (2018). Mental models associated to voice user interfaces for infotainment systems. Dyna 93:245. doi: 10.6036/8766
Ma, L. (1998). The formation and application of speech regulation theory. Hum. Soc. Sci. J. Hainan Univ. 1, 78–81.
Ma, Q., Zhou, R., Zhang, C., and Chen, Z. (2022). Rationally or emotionally: how should voice user interfaces reply to users of different genders considering user experience? Cogn. Tech. Work 24, 233–246. doi: 10.1007/s10111-021-00687-8
Meena, R., Skantze, G., and Gustafson, J. (2014). Data-driven models for timing feedback responses in a map task dialogue system. Comput. Speech Lang. 28, 903–922. doi: 10.1016/j.csl.2014.02.002
Meng, G. (2006). Chinese language speed and listening teaching as a second language. World Chin. Teach. 2, 129–137.
Moore, B. A., and Urakami, J. (2022). The impact of the physical and social embodiment of voice user interfaces on user distraction. Int. J. Human Comput Stud 161:102784. doi: 10.1016/j.ijhcs.2022.102784
Murad, C., Munteanu, C., Cowan, B. R., and Clark, L. (2019). Revolution or evolution? Speech interaction and HCI design guidelines. IEEE Pervasive Comput. 18, 33–45. doi: 10.1109/MPRV.2019.2906991
Myers, C, Furqan, A, Nebolsky, J, Caro, K, and Zhu, J. (2018). Patterns for how users overcome obstacles in voice user interfaces. In Proceedings of the 2018 chi conference on human factors in computing systems (chi 2018), New York, NY, Assoc Computing Machinery.
Myers, C M, Furqan, A, and Zhu, J. (2019). The impact of user characteristics and preferences on performance with an unfamiliar voice user interface, chi 2019. In Proceedings of the 2019 chi conference on human factors in computing systems, New York, NY, Assoc Computing Machinery.
Nass, C, Steuer, J, and Tauber, E R. (1994). Computers are social actors, In Proceedings of the SIGCHI conference on human factors in computing systems, 72–78
Ostrowski, A. K., Fu, J., Zygouras, V., Park, H. W., and Breazeal, C. (2022). Speed dating with voice user interfaces: understanding how families interact and perceive voice user interfaces in a group setting. Front. Robot AI 8:730992. doi: 10.3389/frobt.2021.730992
Page, T. (2014). Touchscreen mobile devices and older adults: a usability study. Int. J. Hum. Fact. Ergon. 3:65. doi: 10.1504/IJHFE.2014.062550
Porcheron, M, Fischer, J E, and Valstar, M. (2020). NottReal: a tool for voice-based wizard of Oz studies, In Proceedings of the 2nd conference on conversational user interfaces, Bilbao Spain: ACM, 1–3.
Portet, F., Vacher, M., Golanski, C., Roux, C., and Meillon, B. (2013). Design and evaluation of a smart home voice interface for the elderly: acceptability and objection aspects. Pers. Ubiquit. Comput. 17, 127–144. doi: 10.1007/s00779-011-0470-5
Powers, A, and Kiesler, S. (2006). The advisor robot: Tracing people's mental model from a robot's physical attributes. In Proceedings of the 1st ACM SIGCHI/SIGART conference on human-robot interaction, New York, NY: Association for Computing Machinery, 218–225.
Pradhan, A., Lazar, A., and Findlater, L. (2020). Use of intelligent voice assistants by older adults with low technology use. ACM Trans. Comput. Hum. Interact. 27:31. doi: 10.1145/3373759
Rakotomalala, F., Randriatsarafara, H. N., and Hajalalaina, A. R. (2021). Voice user Interface: literature review, challenges and future directions. Syst. Theor. Control Comput. J. 1, 65–89. doi: 10.52846/stccj.2021.1.2.26
Rönnberg, J., Arlinger, S., Lyxell, B., and Kinnefors, C. (1989). Visual evoked potentials: relation to adult speechreading and cognitive function. J. Speech Lang. Hear. Res. 32, 725–735. doi: 10.1044/jshr.3204.725
Sanju, K., Himanshu, D. R., and Yadav, A. K. (2019). Relationship between listening, speech and language, cognition and pragmatic skill in children with cochlear implant. IP Indian J. Anat. Surg. Head, Neck Brain 5, 72–75. doi: 10.18231/j.ijashnb.2019.019
Sayago, S, Neves, B, and Cowan, B. (2019). Voice assistants and older people: Some open issues. In Proceedings of the 1st International conference on conversational user interfaces, New York, NY, Association for Computing Machinery, 1–3
Scahill, R. I., Frost, C., Jenkins, R., Whitwell, J. L., Rossor, M. N., and Fox, N. C. (2003). A longitudinal study of brain volume changes in Normal aging using serial registered magnetic resonance imaging. Arch. Neurol. 60, 989–994. doi: 10.1001/archneur.60.7.989
Shahrebaki, AS, Imran, A, Olfati, N, and Svendsen, T. (2018). Acoustic feature comparison for different speaking rates, In proceeding of the human-computer interaction: interaction technologies, HCI International 2018, Part III, Cham, 10903, 176–189.
Shin, A, Oh, J, and Lee, J. (2019). Apprentice of Oz: human in the loop system for conversational robot wizard of Oz, 14th ACM/IEEE International conference on human-robot interaction (HRI), 516–517, 2019.
Singh, M, Rafat, Y, Bhatt, S, Jain, A, and Dev, A. (2021). “Continuous Speech Recognition Technologies-A Review”, in Recent Developments in Acoustics. Singapore: Springer Singapore, 85–94.
Song, Y., Yang, Y., and Cheng, P. (2022). The investigation of adoption of voice-user Interface (VUI) in smart home systems among Chinese older adults. Sensors 22:1614. doi: 10.3390/s22041614
Stigall, Brodrick, Waycott, Jenny, Baker, Steven, and Caine, Kelly. (2020). Older adults’ perception and use of voice user interfaces: a preliminary review of the computing literature. In Proceedings of the 31st Australian conference on human-computer-interaction (Ozchi’19), 423–427. New York: Assoc Computing Machinery
Svennerholm, L., Boström, K., and Jungbjer, B. (1997). Changes in weight and compositions of major membrane components of human brain during the span of adult human life of swedes. Acta Neuropathol. 94, 345–352. doi: 10.1007/s004010050717
Truong, K. P., Poppe, R., and Heylen, D. (2010). A rule-based backchannel prediction model using pitch and pause information. Interspeech, 3058–3061. doi: 10.21437/Interspeech.2010-59
Wang, H. (2002). Changes of physiological function and clinical rational drug use in the elderly. Chin. Commun. Physician 18, 7–8.
White, K. F., and Lutter, W. G. (2003). Behind the curtain: lessons learned from a wizard of Oz field experiment. SIGGROUP Bullet. 24, 129–135. doi: 10.1145/1052829.1052854
Wilkinson, C., and Cornish, K. (2018). An overview of participatory design applied to physical and digital product interaction for older people. Multimodal Technol. Interact. 2:79. doi: 10.3390/mti2040079
Yuan, Y. (1992). A study of speech regulation theory in sociolinguistics. Foreign Lang. Teach. Res., 18–24+80.
Yuan, J, Liberman, M, and Cieri, C. (2006). Towards an integrated understanding of speaking rate in conversation. In Proceeding of Interspeech 2006 and 9th International conference on spoken language processing, 1-5, Baixas: ISCA-INT Speech Communication Assoc, 541–544
Zen, H, Senior, A, and Schuster, M, (2013) Statistical parametric speech synthesis using deep neural networks, IEEE International conference on acoustics, speech and signal processing (ICASSP), New York, 7962–7966
Zhang, Y. (2021). Research on emotional design of speech interaction in the elderly. Audio Eng. 45, 28–30.
Zhang, Z., Huang, M., Zhao, Z., Ji, F., Chen, H., and Zhu, X. (2018). Memory-augmented dialogue management for task-oriented dialogue systems. arXiv. doi: 10.48550/arXiv.1805.00150
Zhang, K., and Zhang, T. (2019). Interface usability for the elderly users in the past 10 years. Packag. Eng. 40, 217–222.
Keywords: voice user interface (VUI), elderly people, feedback speech rate, regulation strategy, speech convergence
Citation: Wang J, Yang S and Xu Z (2023) Talk like me: Exploring the feedback speech rate regulation strategy of the voice user interface for elderly people. Front. Psychol. 14:1119355. doi: 10.3389/fpsyg.2023.1119355
Edited by:
Xin Zhang, Tianjin University, ChinaReviewed by:
Jeffrey Ho, Hong Kong Polytechnic University, Hong Kong, SAR ChinaRaúl Marticorena, University of Burgos, Spain
Sheng Tan, Trinity University, United States
Copyright © 2023 Wang, Yang and Xu. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Junfeng Wang, d2FuZ2p1bmZlbmdAc3p0dS5lZHUuY24=