Pith. sign in

REVIEW 4 major objections 6 minor 32 references

Get It Right: Improving Comprehensibility with Adaptable Speech Expression of a Humanoid Service Robot

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A humanoid service robot can improve comprehensibility of complex official information by offering it in easy language or a foreign language, and the paper shows a working architecture for this with the Pepper robot and external AI…

desk verdict A well-described proof-of-concept for easy-language and translation on a Pepper robot, but the title claims a comprehension benefit that no data in the paper supports. read the letter →

arxiv 2412.05022 v1 pith:6JEXLBHZ submitted 2024-12-06 cs.RO cs.AI

Thomas Sievers , Ralf Moeller This is my paper · ORCID
classification cs.ROcs.AI
keywords socialrobothuman-robotinteractioneasylanguagecomprehensibilitytranslationservicepublicPepper
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper argues that a humanoid service robot can make complex official information easier for customers to understand by automatically offering it in easy language or in a foreign language. The authors propose an application architecture for the Pepper robot that connects to external AI translation services (SUMM AI for easy German, DeepL for other languages) and present a case study in a public-service customer centre. They report that the integration works via standard APIs and that workshops with public-service staff found the features useful. The paper thereby establishes a working integration pattern for language-adaptive robot speech, leaving the effect on customer comprehension as a stated hypothesis for future practical tests.

What carries the argument

The key machinery is the API-based integration pattern: an Android application on Pepper's tablet controls the robot's native speech and listening functions, a PHP-backed MySQL knowledge base returns the original expert text as JSON, and that text is passed through the external AI services SUMM AI (easy language) and DeepL (translation), whose JSON responses are then spoken aloud and shown on the tablet. The dialogue uses Pepper's Chat feature, which the authors report understands words and short sentences even inside longer utterances. This combination lets the robot adapt the level of language difficulty to the individual customer.

What would settle it

A controlled comprehension test with real customers of a public authority, comparing understanding of the same administrative procedure when presented in standard language, easy language, or a translated language by the robot, would settle the claim. If measured comprehension does not improve or worsens for the easy-language or translated conditions, the proposed benefit does not hold. A concrete check: take a set of official procedures, run their texts through SUMM AI and DeepL, and have target-group users rate the accuracy and fluency of the output.

Watch

Extended reading notes

Core claim

The central claim is that the content-related difficulty of a service robot's speech can be adapted on demand: a customer can ask, by voice or via the tablet, for information to be simplified into easy language or translated into another language, and the robot will fetch the simplified or translated text from external AI services and speak and display it. The authors demonstrate this with the Pepper robot, using its Android SDK, a MySQL knowledge base, and API connections to SUMM AI and DeepL. They further report that Pepper's Chat feature, unlike its native Listen action, understands words and short sentences even inside longer utterances, enabling a more natural dialogue. The intended benefit is improved comprehensibility for customers with language or comprehension difficulties, leading to better preparation, less frustration, and more efficient encounters with human case workers.

Load-bearing premise

The claim depends on the untested assumption that SUMM AI and DeepL return simplified or translated text that is accurate and fluent enough for official public-service content, so that the simplified or translated instructions do not introduce new errors that reduce comprehension.

Editorial extensions

If this is right

  • Public-service robots can be deployed with language-adaptation features using off-the-shelf external services, without building the simplification or translation models themselves.
  • Customers who struggle with administrative language can hear and read the same information in easy language or in their own language, which may reduce anxiety and misunderstandings before they meet a human case worker.
  • The pattern is not tied to Pepper: any robot with a programmable interface and API access could use the same architecture.
  • Workshops with public-service staff indicate willingness to use such robots in suitable authorities, supporting practical deployment.
  • Using Pepper's Chat feature rather than its native Listen function improves spoken-language recognition, making natural dialogue with customers more feasible.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper does not measure comprehension outcomes, so a plausible next step is a field study that measures whether customers actually understand the simplified or translated instructions better than the originals; such a study would also reveal whether the simplification process introduces new inaccuracies.
  • The architecture treats text output as the only channel of adaptation; a testable extension is to adapt the robot's speaking rate, pauses, and word emphasis based on the detected difficulty level, which the SDK already supports.
  • Since SUMM AI currently only supports German easy language and the proof-of-concept used Danish for translation, the claimed benefit is geographically limited; extending to other easy-language and target-language pairs would be needed before generalising.
  • The reliance on cloud APIs raises data-protection questions for public-service contexts; the paper mentions data protection as future work, and a local or on-device translation alternative (such as the Google ML Kit cited in related work) could be compared.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper describes a case study and application architecture for a Pepper humanoid robot in a public service setting. The robot retrieves administrative information from a MySQL-backed knowledge base and, on request, re-renders it via SUMM AI into German easy language or via DeepL into a foreign language (Danish in the proof of concept), with output through speech and the robot's tablet. The authors argue that this approach improves comprehensibility, reduces customer uncertainty, and makes service interactions more effective. The paper reports workshops with public service staff and a claimed improvement from using Pepper's Chat feature, but it presents no quantitative evaluation and explicitly defers practical tests of the hypothesis to future work.

Significance. If validated with empirical data, the idea of coupling a social robot to external simplification and translation services is a useful and comparatively low-cost integration pattern for public-service HRI, and the focus on easy language is genuinely under-represented in the HRI literature. The paper also clearly documents a concrete implementation path that other developers could follow. However, as it stands the contribution is an architecture description plus an untested hypothesis; no code, dataset, or behavioral evidence is provided, and the title and abstract make causal claims that the manuscript does not support. The authors' explicit acknowledgment that practical tests are still future work is a positive sign of honesty, but it also confirms that the central claim is unverified.

major comments (4)
  1. [Section 7, first paragraph] The central claim that the proposed method "improve[s] the comprehensibility of complex official information" is not supported by any empirical evidence. There is no user study, no comprehension metric, and no outcome data. The same paragraph states that "practical tests of our hypothesis" are planned for future work, which makes explicit that the hypothesis is currently untested. Please either add an empirical evaluation with a comprehension measure (for example, a between-subjects task with real or simulated customers measuring task success, accuracy of understanding, or self-reported certainty) or revise the title, abstract, and conclusions to present the work as an architecture proposal and a stated hypothesis rather than as a demonstrated improvement.
  2. [Section 6.2] The statement that Pepper's Chat feature "was found to bring a significant improvement in terms of spoken language recognition" is reported without any data, subject pool, or metric. Because this claim partially motivates the architecture, it is load-bearing and should be documented quantitatively, including the number of utterances tested, the recognition accuracy for Listen versus Chat, and the statistical test used. If no formal comparison was performed, please remove the word "significant" and describe the observation more cautiously.
  3. [Section 6, API integration] The manuscript never evaluates the accuracy or fluency of SUMM AI's easy-language output or DeepL's translations for administrative German. The entire comprehension benefit rests on the premise that these external services do not introduce new errors or awkward phrasings that could actually reduce understanding. Please report an evaluation of a representative set of official texts, for example human ratings of comprehensibility and correctness by target-group members, or a systematic error analysis of the simplified and translated outputs.
  4. [Section 7, workshops paragraph] The workshops with public service staff are used as evidence that the proposed features are useful, but no methodology, number of participants, interview protocol, or analysis is reported. Such anecdotal feedback cannot support the claim that customer comprehensibility improves. Please either report workshop data in a systematic way (participant demographics, data collection, coding, findings) or describe the workshops as informal feedback and avoid using them as evidence for the comprehensibility claim.
minor comments (6)
  1. [Title and Abstract] "Improving Comprehensibility" and "improves the intelligibility" overstate the evidence; consider using "proposes" or "aims to improve" until empirical support is available.
  2. [Section 2, paragraph 5] The word "cliens" appears where "clients" is intended.
  3. [Reference [32]] "Wikipedia contributers" should be spelled "Wikipedia contributors" and entry titles should be capitalized consistently.
  4. [Reference [4]] "Havard Business Review Press" should be "Harvard Business Review Press."
  5. [Figures 3, 4, and 5] The screenshots are difficult to read in print; please increase resolution and add callouts or annotations so the reader can see the relevant dialogue and gesture components.
  6. [Reference [28]] Reference [28] is listed as "to be published" and is used to describe the API integration details. If a preprint exists, please provide a stable DOI or arXiv identifier so that the architectural claims can be checked.

Circularity Check

0 steps flagged · score 1.0 of 10

No circular derivation: this is an integration case study whose central claim is untested but not self-referential; the only self-citation is minor and non-load-bearing.

full rationale

The paper is a systems and integration paper, not a derivation. Its central claim, stated in Section 7 ('With this paper we propose a method to improve the comprehensibility of complex official information for human customers by adapting the level of content-related difficulties'), is an architectural proposal: a Pepper robot is connected to SUMM AI for easy language and DeepL for translation, and the output is spoken and displayed on the tablet. There are no equations, fitted parameters, or predicted quantities that are constructed from their own inputs, so the standard circularity patterns such as prediction-equals-fit or self-definition do not apply. The only self-citation is reference [28] (Sievers et al.), which is used to point to a more detailed description of the API integration ('This method and the use of the APIs is described in more detail by Sievers et al. [28]'). That citation supports implementation detail, not the truth of the comprehensibility claim, so it is not load-bearing. The paper is honest that no behavioral validation was performed: Section 7 says 'Through further workshops with authorities and practical tests of our hypothesis in an environment corresponding to our case study, we will seek to gain a better understanding...' The claims about SUMM AI and DeepL output quality, the alleged 'significant improvement' of the Chat feature in Section 6.2, and the effect on customer comprehension are unsupported by data, but lack of empirical evidence is a correctness and validity concern, not circularity. No step in the paper reduces to its own input by definition or through a self-citation chain. The low score reflects only the presence of a minor, non-load-bearing self-citation.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The paper introduces no free parameters or invented entities. It depends on four external assumptions about easy language effectiveness, trust and anthropomorphism, speech recognition quality, and machine-translation accuracy; none of these are validated in the paper.

assumptions (4)
  • domain assumption Easy language improves comprehension for the target groups (people with learning difficulties, dementia, low administrative-language literacy, or second-language learners).
    Invoked in Section 2 (Preliminaries) as the motivation for the entire system, citing refs 12-14, but never tested in the context of robot speech.
  • domain assumption Trust is a key driver of service-robot use, and anthropomorphism increases trust.
    Sections 1 and 4 cite refs 6, 8, 9, 25 to justify choosing a humanoid robot; this is background literature, not demonstrated in this case study.
  • domain assumption Pepper's Chat feature yields sufficiently reliable speech recognition to understand users' requests in a real environment.
    Section 6.2 states 'This was found to bring a significant improvement' but provides no recognition-rate data or test conditions.
  • domain assumption The output of SUMM AI and DeepL is accurate enough for administrative content.
    Section 6 connects to these APIs and treats their output as trustworthy; no quality checks or error analyses are reported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Get It Right: Improving Comprehensibility with Adaptable Speech Expression of a Humanoid Service Robot." pith.science (2026). https://pith.science/paper/6JEXLBHZ

@misc{pith2026241205022,
  author       = {Pith},
  title        = {Pith review of: Get It Right: Improving Comprehensibility with Adaptable Speech Expression of a Humanoid Service Robot},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6JEXLBHZ}},
  note         = {Machine review of arXiv:2412.05022}
}
read the original abstract

As humanoid service robots are becoming more and more perceptible in public service settings for instance as a guide to welcome visitors or to explain a procedure to follow, it is desirable to improve the comprehensibility of complex issues for human customers and to adapt the level of difficulty of the information provided as well as the language used to individual requirements. This work examines a case study using a humanoid social robot Pepper performing support for customers in a public service environment offering advice and information. An application architecture is proposed that improves the intelligibility of the information received by providing the possibility to translate this information into easy language and/or into another spoken language.

Figures

Figures reproduced from arXiv: 2412.05022 by the authors.

Figure 1
Figure 1. Humanoid Robot Pepper [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. Tablet as user interface displaying obtained information translated to Danish the database. This basic example knowledge base could be something totally different in another case of application. It represents the initial information that needs to be simplified. For translation into easy language and into a foreign language, we use services that are hosted on various external servers and are not proprietary. These se… view at source ↗
Figure 3
Figure 3. Main Activity of our application 6.2 Listen & Say or Chat Our first approach was to use the Pepper-specific Listen and Say functionalities [29] [30] to realise a dialogue between robot and customer. The Pepper robot is able to react to words or phrases provided in the program code of the application. Real speech recognition in a general sense is not possible with Pepper’s native abilities but the robot is able to sp… view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Robot emulation viewer showing dialogue between robot and human with knowledge base answer, translation to easy language, and translation to Danish [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: Robot emulation viewer showing the gesture of an animation to generate human-like text may be an option. Data protection concerns also require closer consideration. Through further workshops with authorities and practical tests of our hy￾pothesis in an environment corr…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

32 extracted references · 27 canonical work pages

  1. [1]

    Allmendinger and R

    G. Allmendinger and R. Lombreglia, ”Four strategies for the age of smart services,” Harvard Business Review, vol. 83, pp. 131–143, 2005

    work page 2005

  2. [2]

    Stock-Homburg

    R. Stock-Homburg. (2016) Denn sie wissen nicht, was sie tun - Studie der TU Darmstadt zur Robotisierung von B¨ uro- und Dienstleistungsberufen. Transnational Study series ”Robots@work4.0” [Online]. Available: https://www.tu-darmstadt.d e/universitaet/aktuelles_meldungen/archiv_2/2016/2016quartal4/einzelan sicht_162880.de.jsp

    work page 2016

  3. [3]

    Gardecki, M

    A. Gardecki, M. Podpora, R. Beniak and B. Klin, ”The Pepper humanoid robot in front desk application,” Conference Paper: Progress in Applied Electrical Engineer- ing (PAEE), doi: 10.1109/PAEE.2018.8441069, 2018

    arXiv 2018

  4. [4]

    Diana, ”My Robot gets Me: How social design can make new products more human,” Havard Business Review Press, Boston, MA, ISBN 9781633694422, pp

    C. Diana, ”My Robot gets Me: How social design can make new products more human,” Havard Business Review Press, Boston, MA, ISBN 9781633694422, pp. 218-228, 2021

    work page 2021

  5. [5]

    Andrist, M

    S. Andrist, M. Ziadee, H. Boukaram, B. Mutlu, and M. Sakr, ”Effects of Culture on the Credibility of Robot Speech: A Comparison between English and Arabic,” Improving Comprehensibility with adaptable Speech Expression 13 Proceedings of the Tenth Annual ACM/IEEE International Conference on Human- Robot Interaction (HRI ’15), Association for Computing Machi...

    arXiv

  6. [6]

    van Pinxteren, R.W.H

    M.M.E. van Pinxteren, R.W.H. Wetzels, J. R¨ uger, M. Pluymaekers and M. Wetzels, ”Trust in humanoid robots: implications for services marketing“, Journal of Services Marketing 33/4, pp. 507–518, 2019

    work page 2019

  7. [7]

    Mayer, J.H

    R.C. Mayer, J.H. Davis and F.D. Schoorman, ”An integrative model of organi- zational trust“, Academy of Management Review, Vol. 20 No. 3, pp. 709–734, https://doi.org/10.2307/258792, 1995

    doi:10.2307/258792 1995

  8. [8]

    Everett, D

    J. Everett, D. Pizarro and M. Crockett. (2017) Why are we reluctant to trust robots? [Online]. Available: https://www.theguardian.com/science/head-quarters/20 17/apr/24/why-are-we-reluctantto-trust-robots

    work page 2017

Show all 32 references
  1. [9]

    B. Morgan. (2017) 10 Things robots can’t do better than humans [Online]. Available: https://www.forbes.com/sites/blakemorgan/2017/08/16/10-things-robots-c ant-do-better-than-humans

    2017

  2. [10]

    Epley, A

    N. Epley, A. Waytz and J.T. Cacioppo, ”On seeing human: a three-factor theory of anthropomorphism“, Psychological Review, Vol. 114 No. 4, pp. 864–886, https: //doi.org/10.1037/0033-295X.114.4.864, 2007

    2007 doi

  3. [11]

    Furlough, T

    C. Furlough, T. Stokes and DJ. Gillan, ”Attributing Blame to Robots: I. The Influence of Robot Autonomy,” Hum Factors, Vol. 63 No. 4, pp. 592–602, https: //doi.org/10.1177/0018720819880641, 2021

    2021 doi

  4. [12]

    (2021) Any complex content can be translated into easy language [Online]

    Press and Information Office of the Federal Government. (2021) Any complex content can be translated into easy language [Online]. Avail- able: https://www.bundesregierung.de/bregde/aktuelles/interview-anne-leichtfuss- 1918176

    2021

  5. [13]

    Quatropus GmbH & Co. KG. (2022) On the status quo of easy language in Germany [Online]. Available: https://www.quatrolingo.com/status-quo-leichte- sprachedeutschland-2022/

    2022

  6. [14]

    Easy-to-understand language and barrier-free information [Online]

    capito. Easy-to-understand language and barrier-free information [Online]. Avail- able: https://www.capito.eu/en/what-is-easy-to-understand-language/

  7. [15]

    Driving Sales and Satisfaction with Multilingual Service [On- line]

    Michael Soon Lee. Driving Sales and Satisfaction with Multilingual Service [On- line]. Available: http://resources.rosettastone.com/CDN/us/pdfs/Biz-Publi c-Sec/Driving-Sales-and-Satisfaction-with-Multilingual-Service.pdf

  8. [16]

    Initiative D21 e. V. (2022) eGovernment MONITOR [Online]. Available: https: //initiatived21.de/egovernment-monitor/

    2022

  9. [17]

    Shin, E.T

    S. Shin, E.T. Matson, Jinok Park, Bowon Yang, Juhee Lee and Jin-Woo Jung, ”Speech-to-speech translation humanoid robot in doctor’s office,” 2015 6th Interna- tional Conference on Automation, Robotics and Applications (ICARA), pp. 484– 489, https://doi.org/10.1109/ICARA.2015.70...

    2015

  10. [18]

    Leeuwestein, M

    H. Leeuwestein, M. Barking, H. Sodacı et al., ”Teaching Turkish-Dutch kinder- gartners Dutch vocabulary with a social robot: Does the robot’s use of Turkish translations benefit children’s Dutch vocabulary learning?” Journal of Computer Assisted Learning, 37, pp. 603–620, http...

    doi

  11. [19]

    Santano, innovex GmbH (2023) How to Use Google’s ML Kit to Enhance Pepper With AI (Part 5) [Online]

    S. Santano, innovex GmbH (2023) How to Use Google’s ML Kit to Enhance Pepper With AI (Part 5) [Online]. Available: https://www.inovex.de/de/blog/how-to-use- googles-ml-kit-to-enhance-pepper-with-ai-part-5/

    2023

  12. [20]

    (2023) Translation [Online]

    Google. (2023) Translation [Online]. Available: https://developers.google.com/ml- kit/language/translation

    2023

  13. [21]

    Lastrico, N

    L. Lastrico, N. F. Duarte, A. Carf ` ı, F. Rea, F. Mastrogiovanni, A. Sciutti and J. Santos-Victor, ”If You Are Careful, So Am I! How Robot Communicative Motions 14 Thomas Sievers and Ralf M¨ oller Can Influence Human Approach in a Joint Task,” https://doi.org/10.48550/ARX IV....

    doi arXiv

  14. [22]

    van Deurzen, H

    B. van Deurzen, H. Bruyninckx and K. Luyten, ”Choreobot: A Reference Frame- work and Online Visual Dashboard for Supporting the Design of Intelligible Robotic Systems,” Proc. ACM Hum.-Comput. Interact. 6, EICS, Article 151, 24 pages. https://doi.org/10.1145/3532201, 2022

    doi

  15. [23]

    (2022) Pepper [Online]

    Aldebaran, United Robotics Group and Softbank Robotics. (2022) Pepper [Online]. Available: https://www.aldebaran.com/en/pepper

    2022

  16. [24]

    (2022) Pepper SDK for Android [Online]

    Aldebaran, United Robotics Group and Softbank Robotics. (2022) Pepper SDK for Android [Online]. Available: https://qisdk.softbankrobotics.com/sdk/doc /pepper-sdk/index.html

    2022

  17. [25]

    J. Fink, ”Anthropomorphism and Human Likeness in the Design of Robots and Human-Robot Interaction,” in International Conference on Social Robotics (ICSR), Chengdu, China, Springer, Heidelberg, Berlin, pp. 199–208, https://doi.org/10 .1007/978-3-642-34103-8_20 , 2012

    2012

  18. [26]

    (2022) Summ - easy language [Online]

    SUMM AI GmbH. (2022) Summ - easy language [Online]. Available: https://summ-ai.com/en/

    2022

  19. [27]

    (2022) Translate with the deepl api [Online]

    DeepL SE. (2022) Translate with the deepl api [Online]. Available: https://www.deepl.com/pro-api

    2022

  20. [28]

    Sievers, M

    T. Sievers, M. Bender and R. M¨ oller, ”Connecting AI Technologies as Online Ser- vices to a Humanoid Service Robot,” to be published in: 15th International Confer- ence on Computer and Automation Engineering, (ICCAE 2023), March 3-5 2023

    2023

  21. [29]

    (2022) Listen [Online]

    QiSDK. (2022) Listen [Online]. Available: https://qisdk.softbankrobotics.c om/sdk/doc/pepper-sdk/ch4_api/conversation/reference/listen.html

    2022

  22. [30]

    (2022) Say [Online]

    QiSDK. (2022) Say [Online]. Available: https://qisdk.softbankrobotics.com/s dk/doc/pepper-sdk/ch4_api/conversation/reference/say.html

    2022

  23. [31]

    (2022) Chat [Online]

    QiSDK. (2022) Chat [Online]. Available: https://qisdk.softbankrobotics.com /sdk/doc/pepper-sdk/ch4_api/conversation/reference/chat.html

    2022

  24. [32]

    (2022) Uncanny valley [Online]

    Wikipedia contributers. (2022) Uncanny valley [Online]. Available: https://en.w ikipedia.org/wiki/Uncanny_valley

    2022

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.