September 2019
Speak it is an application that allows people with deafness, cerebral palsy, ALS, etc. communicate with those around you. For this, it has several communication methods:
– Chat – Allows you to convert written text to speech and convert spoken text to text.
– Pictograms – Allows you to verbalize feelings, body parts, actions, places, etc. through the use of pictograms that are associated with a text that indicates what they symbolize.
– Shortcuts – Allows you to quickly speak frequently used phrases.
Next, an analysis of the different functions of the application will be carried out.
Features
The application has three functions, the Chat, the Pictograms and the Shortcuts. These functions are accessed from the beginning (image on the left) or from the application menu (image on the right).
Next, an analysis of the different functions that the application has will be carried out.
Cat
This function allows you to have a conversation with a person by writing text and converting it into voice, as well as transforming the voice into written text. To do this, the chat has two methods of entering text:
- Talk. When the talk button is pressed, a keyboard opens for typing text. Once the text has been entered, the send button is pressed, located to the right of the editing box, so that the text is spoken aloud and is reflected in the chat, as can be seen in the image on the left .
- Hear. When this button is pressed, a box is displayed with a button that must be pressed and held while speaking. After releasing it, you have to press the button located to the right of the text that has been written so that it is reflected in the chat, as can be seen in the image on the right.
Pictograms
This option allows communication through the use of pictograms. These are grouped by categories.
The structure of the pictograms is in the form of a grid, each row being a category of pictograms, and each pictogram occupies a cell within the structure.
Scrolling through the pictograms is done by sliding your finger on the screen. If the movement is lateral, it changes between pictograms of the same category. If the scrolling is vertical, it switches between the different categories.
The application allows the modification and elimination of the pictograms, although it does not allow their relocation. It is also possible to create new pictograms within the same category, which are placed at the end of it.
Shortcuts
This function gives access to frequently used phrases, related to the person such as the name or age, to feelings such as happy or sad, actions such as cooking or playing and places such as "Where is the nearest hospital" or "Where is a policeman?". You can also add your own phrases that are often used.
Configuration
The application allows you to configure different parameters of it. These are grouped into 4 sections:
- General. These are aspects related to the appearance such as the size of the text and the glyphs, if the text is shown in upper case, if vibration is used and if a dark theme is used in the application.
- Profile. They are parameters related to the user such as language, name, date of birth, gender and if emergency data is displayed.
– Text to speech. These are parameters related to the voice of Google synthesis such as speed and pitch.
– Voice to text. They are parameters related to voice recognition by Google.
Tests performed
In order to analyze and be able to carry out an evaluation of the application, a series of tests have been carried out to verify the behavior of the different functions of the application. In this sense, the general operation of the application is satisfactory, although with certain aspects that will be discussed below.
– Chat. The tests carried out in order to evaluate this functionality have provided very satisfactory results. The voice recognition used to display the text of a speaking party on the screen has worked perfectly, although this is not something specific to the application but rather a Google service. For its part, the speech system through Google's voice synthesis has two drawbacks: on the one hand, the voice synthesis used by another that the user has installed on their device cannot be modified, and on the other, the speech synthesis does not differentiate intonation between question, exclamation and neutral text, so without reading the text it is not known if it is a question, a neutral text or a sentence with exclamation.
– Pictograms. In the tests carried out in this function, it is where the greatest number of aspects to be improved have been detected. Although it is true that the pictograms are well categorized, with the most used being represented and with a good number of categories, and that the user can add more, these should be able to be ordered, in such a way that the user places them at the beginning of the category the ones you use most often. In addition, scrolling through the pictograms and their categories is too slow, and a faster mechanism should be implemented, such as the inclusion of scroll bars or the possibility of enlarging or reducing the icons to have an overview of them. It should also be mentioned that, as previously indicated, pictograms can be added to the categories, which are located at the end of the category, although it would be more logical to place them at the beginning of the category or that the user could choose the position within the category. category. Finally, mention that in the application settings you can modify the size of the pictograms, although the smallest size of the pictograms is still too large.
– Shortcuts. The shortcuts are fine, although in the tests carried out, Google's speech synthesis has not correctly intoned the questions. Although it is not an application development but a Google service, it is very important to review it here so that the user is aware of this fact. Perhaps, if the voice synthesis could be changed to a third-party one, this aspect would be resolved. In everything else, the shortcuts have worked correctly.
Finally we will mention that the application has some elements, especially buttons, that are not labeled. This could be useful for deafblind people who want to use the app. Also indicate that the dark theme can be activated, which facilitates the use of the application by people with low vision.
Conclusion
The application generally works well, helping people with disabilities who cannot communicate verbally to establish conversations with people who do not have speech difficulties.
The chat function is undoubtedly the strong point of the application, allowing to establish a dialogue between a person with speech problems and another who does not.
The less positive aspects of the application are related to the pictograms, which should be improved for a faster and more comfortable scrolling through them and their categories, and allowing the reordering of them, in such a way as to reduce as much as possible the time to find the desired pictogram.
Finally, it should be noted that the fact of being a free application is a very favorable point to consider this application as a communication system for people with hearing problems or other pathologies such as cerebral palsy or ALS. the only downside in this regard is that it is only available for Android users since there is no version for IOS, which would be something very interesting.
Highlights
- It allows the conversation between a person with communication problems and another without them.
- Allows voice recognition.
- Convert text to speech.
- Well categorized and intuitive pictograms.
- Shortcuts to frequently used phrases.
- Is free.
Improvement points
- One could suggest the possibility of including faster scrolling mechanisms through the pictograms and their categories.
- The possibility of allowing the reordering of the categories and the pictograms within them in future versions could be studied.
- It would be interesting if the speech synthesis could be changed.
- The possibility of developing a version for IOS could be studied.