Is this actually, 100% free?

Yes! It's for those who can't afford an iPad and apps, or can't afford or use the computer software available. I want people with aphasia to be able to practise at home with more than pen and paper, without needing assistance from carers and without spending money. I don't make any profit or take any payment and volunteer the coding and time. You might like to donate a few dollars if it has helped someone you know ☕️.

Therapy Tasks

• Letter matching across case and font
• Picture to written word matching (either multiple pics or multiple words)
• Picture to auditory word matching
• Written to auditory word matching
• Written to auditory letter matching
• Typing spoken words
• Typing spoken letters
• Written picture naming
• Spoken picture naming
• Answering written yes/no questions – simple, passive and comparative

Site Features

• Multiple languages! Read more about them below.
• Wordlists based on difficulty (easy through hard) or semantic categories
• The hard lists really are hard!
• To save settings, copy the web address - the link contains the settings so you can copy or email the link to your patient!
• Two images for each word to avoid shallow word-pic association (randomised on each presentation)
• Animated gifs for verbs (a few)
• Written naming tasks accept multiple plausible responses
• Only one attempt permitted in matching tasks. This means the patient can't just keep guessing until they get the right answer, and avoids reinforcement of the wrong response.
• Cues for all tasks (Called 'Hints' for simplicity)

Rationales for Setup

How did you choose target words?

Generalisation of treated items to conversation has been demonstrated in studies^{(e.g., 1 2 3)}, but evidence for generalisation of treated items to untreated items is not promising^{(e.g. 4 5)}. It is therefore important that words chosen for therapy are highly useful¹. The words for this website are chosen in order of frequency to be most functional. This was, of course, limited by what words can accurately be represented by images.

Each wordlist has approximately 40-50 words (see table below). There are some studies suggesting that even people with severe aphasia can benefit from a high number of treatment words⁶ so I saw no need to limit word lists to less than this.

For English, I used the SUBTLEXUS corpus which contains verbal frequency ratings for words. While it is based on subtitles and is therefore only 'TV' vocabulary, it is likely more representative of real verbal communication than written frequency. There were some items with high frequency (e.g. 'bomb') that I excluded as they are [hopefully] not often used in everyday language. For colour, clothing and numbers I added relevant words myself, and for gardening I retrieved the highest frequency words from the full text of a gardening book. For other languages, words were largely translated from the English words after excluding culturally irrelevant concepts and wordlists were regrouped according to frequency data for each language (See details).

Cueing Hierarchy (naming task)

An increasing cueing hierarchy is used. Increasing and decreasing cues appear to be equally efficacious ^{(7 8 9)}. The cues used (semantic → sentence → letter → phonemic → written) are based loosely on Abel et al. (2005) and general clinical practice.

Details for Each Language

Arabic/‏العربية‎ (2020)

Translation was completed to Arabic, in Saudi Khaleeji dialect, by three thorough and dedicated Speech Language Pathologists in Saudi Arabia.

Frequency derived from the Open Source Arabic Corpora (OSAC) by Saad & Ashour (2010).

English (2013)

Vocabulary and audio are specific to three English-speaking countries – Australia (my location), the USA and England. For example, the same image is depicted as ‘lollies’, ‘candy’ or ‘sweets’ depending on the region. The region should default to your location automatically. I am happy to hear from people interested in adapting materials to other English-speaking countries!

German/Deutsch (2024)

The German translation was carried out as part of a master´s thesis. Lena Werner, speech pathologist, translated the website and adapted it for the German context. As part of the project, the website was also tested with individual people with aphasia. The project was supervised by Dorothea Peitz (M.Sc.) and Dr. Katja Hußmann, who also contributed to and supported the translation and testing.

Frequency derived from the German data of the open subtitle corpus (http://www.opensubtitles.org/).

Greek/Ελληνικά (2024)

The Greek adaptation of aphasia therapy online was generously supported by a Telematics Trust Innovation grant. Dr. Maria Kambanaros, a professor of speech pathology at UniSA and Cyprus University of Technology, oversaw the cultural and linguistic adaptation of the materials.

Greek frequency data comes from Ktori et al. (2008)

Mandarin/中文 (2023)

Mandarin adaptation was completed thanks to the extensive labour of:

• Sajlia Jalil, who led the team from the speech pathology department at Changi General Hospital
• Denise Lee and Chai Cai Ling, speech therapists
• The Telematics Trust also supported this work through an innovation grant.

Wayne Huang, speech pathologist, provided assistance with phonemic cues.

Frequency data was sourced from Cai & Brysbaert (2010) which is drawn from subtitles.

Portuguese/Português

European (2020)

Translation completed with a lot of labour by José Grincho, while Speech Pathologist Luísa Rezende, Occupational Therapist Ana Tavares and Professor José Fonseca provided support and advice.

Frequency data is specific to European Portuguese, from www.linguateca.pt

Brazilian (2021)

Dr Roxele Ribeiro Lima (aphasia researcher) and Marina Antoniazzi Mancini (speech pathologist) volunteered many hours of time to adapt into Brazilian Portuguese and record audio.

Spanish/Español

European (2021)

Thanks to Aitana Viejo Sánchez, who translated the site into Spanish single-handedly! Aitana is a speech pathologist and is working on her PhD at the University of Oviedo, Spain. Her research focuses on pragmatics and social communication, and one of her areas of interest is aphasia.

Frequency information from SUBTLEX-ESP (Cuetos, Glez-Nosti, Barbon & Brysbaert, 2011)

Latin American (2023)

Yoel Droguett and Dr Yina Quique kindly put a lot of work into adapting the Spanish materials for the Latin American context, revising all of the word and sentence data and re-recording all audio files 🙏

Yoel is a Speech-language pathologist from Chile, he works with swallowing, voice and acquired language disorders in long-term care and acute care in Yukon, Canada.

Dr. Quique is a speech and language pathologist from Colombia. She is committed to improving rehabilitation outcomes and healthcare services for stroke survivors with aphasia, especially for Spanish speakers with aphasia.

Turkish/Türkçe (2023)

Many thanks to Mr Memik Yıldız and Dr. Özlem Oğuz for their exceptional work in translating the website into Turkish. Their efforts and commitment have made it possible for Turkish speakers to benefit from the site.

Frequency data comes from a large corpus of word frequency from subtitles: Frequency counts for 4.3m tokens calculated from 824.2m tokens.

Urdu/‏اُردُو‎ (2022)

Alishba Ali and Associate Professor Rashid Mahmood worked hard to translate the website into Urdu - a huge thank you to both for their work. Alishba is a speech pathologist with experience working with aphasia patients at Sheikh Zayed Hospital, Lahore, Pakistan. She is also the National Deputy Head of the Research Department at SYNCH. Associate Professor Rashid Mahmood is a linguistics researcher who has been at Bisha University since 2013. He is the author of 35 research papers in World Englishes, Language Variation, Critical Discourse Analysis, and Stylistics.

Vietnamese/Tiếng Việt (2024)

In Progress!

The Vietnamese adaptation was wholly supported by a Telematics Trust grant. MyTien Lam Lazzarotto provided high quality cultural and linguistic translation services and Genevieve Vuong advised on clinical and technical aspects of adaptation to Vietnamese.

Thank you also to Mr. Tran Huu Duc and Ms. Pham Dinh Ngan Thanh (Lecturers, Faculty of Nursing and Medical Technology, University of Medicine and Pharmacy at Ho Chi Minh city) for advice and feedback.

Frequently Asked Questions

Can I help translate?

Eventually, I would like to offer the most common twenty or thirty languages (or more!) as well as some of the small or endangered languages. However, each one takes a lot of time and effort for volunteers and for me – the site uses nearly 600 target words, and each needs checking for cultural relevance, a translation, a sentence and semantic cue, a phonemic cue, a list of homophones and other appropriate responses to the pictures. Then, there are the hundreds of audio recordings and finally translation of the yes/no questions.

If you think you can manage all this work, I would be happy to hear from you. I prefer to have a team of translators for each language rather than one person. Each language also needs a speech pathologist who is a native speaker and experienced with aphasia.

Can I record a different region/accent for an existing language?

Yes please!

Can the picture naming task ‘hear’ my patient?

No, this is the only task that can’t give feedback on responses. Open source speech recognition is becoming more widely available, but it is less accurate for people with impaired speech or language.

Can I download the yes/no questions?

Yes. There are simple yes/no questions (some reversible, some not), comparative questions (e.g. Is x bigger than y), and passive questions.

Can I download the wordlists and pictures?

Yes, go ahead. The images are all public domain. Friendly reminder though - while there is nothing wrong with practising general language and vocabulary, your patient should ideally have personalised wordlists as well. These should have the most relevant words for their life.