What are the exercises like, and what do they cover?

The therapy exercises in Cuespeak address the full range of difficulties encountered in aphasia and apraxia of speech, including word finding, sentence production, understanding speech, reading comprehension, writing and articulation. Exercises are designed to be enjoyable to carry out, with smatterings of gentle humour here and there.

Most exercises use an interactive question and answer format, meaning that speech and language stimulation takes place in a context which goes some way towards emulating the demands of everyday communication. There is a range of help available for when the questions prove difficult to answer, including spoken feedback on errors and spoken cues to help elicit spoken responses to questions. Exercises are highly customisable to individual needs, making them suitable for people with anything from mild difficulties to very severe difficulties.

Questions & Answers

The main word-finding module in Cuespeak is called QA (short for “Questions and Answers”). QA contains over 1800 questions, which combine two related tasks: a semantic task and a word-finding task. Any element on the screen can be hidden, creating 13 variants (available in settings), each variant placing a different amount of emphasis on the various aspects of language processing, such as understanding spoken questions, reading comprehension, word-finding and articulation, as well as non-verbal semantic difficulties.

Since virtually all people with moderate to severe aphasia have difficulties in at least one of these areas, QA tends to be suitable for most users, and is usually the best starting point in a therapy programme, even if the person’s problems aren’t primarily semantic in nature.

As in many of the Cuespeak modules, every question in QA has a button which, when pressed, displays videos showing the articulation of the target word and the word broken down into syllables and the individual sounds.

Sentence Sorter

Sentence Sorter addresses spoken production of sentences, offering a range of options for how to do this.

In its default state (the sentence ordering task) the user hears/sees a question and, below, sees an answer in the form of a sentence anagram which needs to be arranged into the correct order. To make the task easier you can pre-fill specified tiles in the answer. You can even pre-fill all the words so that the task merely requires reading a sentence aloud in response to a question.

To make the task harder you can hide various elements such as the question, the answer and the picture. You can control the number of words in the answer and the complexity of those words from the point of view of spoken production.


Senfill (short for “sentence fill”) addresses word-finding at the sentence level, the main focus being on verbs.

The user hears or sees a question, and needs to touch the written word that answers that question. The selected word will appear in the context of a spoken and written sentence below. The user then needs to judge whether or not the choice was correct, by touching the check button (the green smiley face). Spoken feedback is given on errors.

As with most exercise modules in Cuespeak, videos are available to help the user to say the target word.


As its name suggests, Soundspell is a spelling task which uses units of sound rather than individual letters. For example, ch is treated as a single unit, since it represents only one sound. This format aims to re-familiarise users with the relationship between word structure and sound, since many people with aphasia lose some or all of this ability. It can serve as an effective facilitator for spoken word production, and articulation videos are included to this end.

The sound/letter tiles can also be presented blank, making a sound sequencing task.


Perspectives is a sentence processing module. This is aimed at anyone who has difficulties understanding or producing sentences.

The main task in Perspectives does not actually require the user to produce a sentence, although it can be used for this. Research has demonstrated that exercises involving judgements about how thematic roles relate to syntactic roles in a sentence (‘mapping’) can have positive benefits for the ability to produce sentences, even when the therapy does not itself require any language output (e.g. E.V. Jones 1986).

Rhyme Judgement

We often focus on a person’s ability to say the word without considering the extent to which they have accessed the sound of the word internally. The integrity of this inner speech as it is sometimes called is an important consideration when diagnosing and treating apraxia of speech.

Asking & Offering

Asking and Offering is designed for practising word retrieval in the context of phrases and sentences that mirror those occurring in everyday conversation.


Letterspell is a more conventional spelling task – no sounds are heard when the tiles are touched. A point is scored for each correct letter, yielding a sensitive score.

Syllable Sorter

Syllable sorter is identical to Soundspell (see above) except that the units here are syllables rather than phonemes.

How Many Letters

Many people with aphasia have difficulty spelling words but can nevertheless retrieve some information about their written form, such as the number of letters. Focussing on this preserved knowledge can serve as a facilitator for word retrieval in both spoken form and written form. How Many Letters is designed to do this. Articulation videos are available for all target words.