The DOI matches its Crossref record; 4 of the 4 references it cites were found in a registry or an index; none of the 2 sources its claims cite is checked for support yet.
The feasibility of eyes-free touchscreen keyboard typing
Abstract
Typing on a touchscreen keyboard is very difficult without being able to see the keyboard. We propose a new approach in which users imagine a Qwerty keyboard somewhere on the device and tap out an entire sentence without any visual reference to the keyboard and without intermediate feedback about the letters or words typed. To demonstrate the feasibility of our approach, we developed an algorithm that decodes blind touchscreen typing with a character error rate of 18.5%. Our decoder currently uses three components: a model of the keyboard topology and tap variability, a point transformation algorithm, and a long-span statistical language model. Our initial results demonstrate that our proposed method provides fast entry rates and promising error rates. On one-third of the sentences, novices' highly noisy input was successfully decoded with no errors.
Overview
CoChat Summary
TL;DR
The paper proposes eyes-free touchscreen typing where users tap out an entire sentence on an imagined Qwerty keyboard without visual reference or intermediate feedback, then an offline decoder recognizes the intended sentence. In two blindfolded experiments, a combined model achieved 18.5% character error rate in the second experiment, with mean entry rates of 29.4 wpm and 23.3 wpm.
Key findings
- In the first experiment, full decoding with transformation and keyboard variance optimization combined gave CER 32.9%, WER 49.1%, and SER 77.8%; mean entry rate was 29.4 wpm.
- In the second experiment, after adding right-swipe spaces and audio feedback, the combination model achieved CER 18.5%, WER 30.1%, and SER 67.9%; mean entry rate was 23.3 wpm.
- The abstract reports that on one-third of sentences, novices’ highly noisy input was successfully decoded with no errors.
- Individual error rates varied; the best user had an error rate of 9.8%, though the paper says why is unknown.
Methods
- First experiment: 14 participants entered 20 randomly chosen short memorable sentences from the Enron mobile test set while blindfolded; no audio or tactile feedback beyond sentence playback, and no error correction.
- Second experiment: 8 participants entered 40 sentences while blindfolded; the app required a right swipe for spaces (touch width over 52 pixels) and added iPhone keyboard click/unlock sounds.
- Decoder components: a keyboard model placing a 2D Gaussian with diagonal covariance on each key, a point transformation allowing x/y scaling, translation, rotation up to 20 degrees, and variance factors, and a 9-gram character language model with Witten-Bell smoothing trained on billions of words of Twitter, Usenet, and blog data (9.8M parameters, 67 MB compressed). Recognition was offline, using grid search over transforms and beam-width pruning for full decoding.
Limitations
- Stated: error rates remain somewhat high; future work includes improving recognition accuracy, implementing efficient error correction interfaces, obtaining a better signal from users, and collecting data from users who are blind or visually impaired.
- Evident from the text: the experiments used small samples (14 and 8 participants), all participants were familiar with Qwerty and were blindfolded rather than blind or visually impaired, and the decoder operated offline.
- Evident from the text: the first experiment provided no mechanism to correct errors and no intermediate letter- or word-level feedback.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 1 Not valid · 1 Not checked · 12 Own results and interpretations
A result the paper states — it cites no source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Participants' mean entry rate was 29.4 wpm.”
- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Combining all models improved accuracy.”
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Our best user had an error rate of 9.8%.”
plainSingle case- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Participants' mean entry rate was 23.3 wpm.”
plainn = 8- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“Independent word transforms did worse on average than a single sentence transform but did help improve accuracy when combined with other models.”
plainn = 8- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
2 more claims
Create a free account to see all 14 claims and every source verdict.
- Borderline: one of the tests for a substantial claim is unclear. 1
- The quote only partly states this claim. 2
- Probably not a substantial claim. 2
- It may state two separate claims as one. 2
Citations
Cites
A versatile dataset for text entry evaluations based on genuine mobile emails
Vertanen, K., & Kristensson, P. O. (2011) · Proceedings of the 13th International Conference on Human Computer Interaction with Mobile Devices and Services · doi:10.1145/2037373.2037418 · Cited by 177
Self-citationAn evaluation of BrailleTouch: mobile touchscreen text entry for the visually impaired
Southern, C., Clawson, J., Frey, B., Abowd, G., & Romero, M. (2012) · Proceedings of the 14th international conference on Human-computer interaction with mobile devices and services · doi:10.1145/2371574.2371623 · Cited by 125
(2012) · Graphics Interface · Cited by 103
Relaxing stylus typing precision by geometric pattern matching
Kristensson, P.-O., & Zhai, S. (2005) · Proceedings of the 10th international conference on Intelligent user interfaces · doi:10.1145/1040830.1040867 · Cited by 88
Self-citation
Research works printed but not linked: 1
- Azenkot S., 2012, Proc. Graphics Interface, 121
Cited by
BlindType: Eyes-Free Text Entry on Handheld Touchpad by Leveraging Thumb's Muscle Memory
Lu, Y., Yu, C., Yi, X., Shi, Y., & Zhao, S. (2017) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3090083 · Cited by 76
i'sFree: Eyes-Free Gesture Typing via a Touch-Enabled Remote Control
Zhu, S., Zheng, J., Zhai, S., & Bi, X. (2019) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3290605.3300678 · Cited by 40
Investigating Gesture Typing for Indirect Touch
Yang, Z., Yu, C., Yi, X., & Shi, Y. (2019) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3351275 · Cited by 23
Typing on Split Keyboards with Peripheral Vision
Lu, Y., Yu, C., Fan, S., Bi, X., & Shi, Y. (2019) · CHI Conference on Human Factors in Computing Systems · doi:10.1145/3290605.3300430 · Cited by 11
Clawson, J., Arif, A. S., Brewster, S. A., Dunlop, M. D., Kristensson, P. O., & Oulasvirta, A. (2015) · CHI Conference Extended Abstracts on Human Factors in Computing Systems (CHI EA) · doi:10.1145/2702613.2702660 · Cited by 7
Self-citation
Metadata
- Authors
- Vertanen, K., Memmi, H., & Kristensson, P. O.
- Affiliations
- Montana Technological University (US)
- University of St Andrews (GB)
- Venue
- Proceedings of the 15th International ACM SIGACCESS Conference on Computers and Accessibility
- Publisher
- ACM
- Published
- 2013-10-21
- Pages
- 1-2
- OpenAlex
- W2012960583
- Cited by
- 13
- References
- 4
- Citations by year
- 2024: 1 · 2023: 1 · 2022: 1 · 2021: 1 · 2019: 4 · 2017: 2 · 2015: 2 · 2014: 1
- FWCI
- 1.59
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- International ACM SIGACCESS Conference on Computers and Accessibility (ASSETS) (publishedVersion)
- Cambridge University Engineering Department Publications Database (submittedVersion)
- Topics
- Life Sciences › Neuroscience › Cognitive Neuroscience › Tactile and Sensory Interactions
- Physical Sciences › Computer Science › Information Systems › User Authentication and Security Systems
- Physical Sciences › Computer Science › Human-Computer Interaction › Gaze Tracking and Assistive Technology
- Keywords
- touchscreen keyboards, QWERTY keyboard, character recognition, point transformations, statistical language models, assistive technology
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: The feasibility of eyes-free touchscreen keyboard typing. https://app.cochat.ai/papers/10.1145/2513383.2513399
- The paper
- Vertanen, K., Memmi, H., & Kristensson, P. O. (2013). The feasibility of eyes-free touchscreen keyboard typing. Proceedings of the 15th International ACM SIGACCESS Conference on Computers and Accessibility. https://doi.org/10.1145/2513383.2513399
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.