Read Aloud is one of the features of Microsoft Immersive Reader tool. This tool provides various possibilities to increase users’ skills in reading texts, comprehension, improving spelling and pronunciation of words. Using Read Aloud as part of this tool is text-to-speech, and Redmondians have made it available in software such as Office, Edge Browser, and Azure Platform.
According to Microsoft, the Read Aloud tool has been developed with the help of several artificial intelligence models, and thanks to neural processing technology, it is possible to convert texts to text with relatively high accuracy in different languages. This feature has supported a limited number of languages so far, however Microsoft announced about a month ago that it would add more language support to the tool, and now Farsi language Is among the supported languages. Farsi-speaking users can use Read Aloud capabilities in various Microsoft software and platforms. According to Microsoft, support for new languages is gradually being added to all Read Aloud-enabled applications.
We have just learned that Persian speech has been activated in the Read Aloud section of the Edge browser, and if you are a user of this browser, you can listen to it by activating it instead of reading web content, which can be very useful and useful. Microsoft uses two Persian-language audio models called Delara (female) and Farid (male); Of course, the Delara audio model is currently available in the Edge browser.
To enable the Read Aloud feature, click the three-dot option at the bottom of the main bar and click Read Aloud, then in the new bar that appears at the bottom of the main bar, click the Voice Options option, and from the Choose a voice section, select Microsoft Dilara Online Select (Natural) – Persian (Iran).
In Zomit experiments, Read Aloud performance is evaluated very well due to the phonetic and phonological complexities of Persian language. It goes without saying that there are still many errors and mistakes in the pronunciation of some words; For example, vowels (short and long sounds of non-Arabic words), some words are mispronounced and numbers are ignored; Of course, the general meaning of the sentences is well understood, and the recognition of punctuation marks is one of the strengths of Read Aloud. If in a Persian text, the punctuation marks are used in the right place, the speaker; Recognizes the pause in the sentence or the question of the sentence and the tone of a phrase in general, and uses the correct vowels, thanks to machine learning and according to the other words used in the sentence, without the need to Arabicize the words.
In addition to the Edge browser, the Read Aloud feature has been added to the web-based version of Microsoft Office, and we tested this feature in web-based software. However, in some cases it still does not work properly in recognizing the language of some Word files and recognizes it as an unsupported language. All you have to do is save your document in VanDrive, then go to the web version of VanDrive, open the document and click on the Editing tab, and in the menu that opens, change the status of the Word file to Viewing; Click on Immersive Reader in the main bar. In the new page that opens, click the play option to have the speaker start reading.
The Read Aloud environment gives you other possibilities; For example, you can change the speaker speed or change the gender of the speaker; Of course, Farid’s audio model does not seem to be complete yet; Because it does not pronounce words correctly compared to Delara. In addition, other features such as changing the background color, changing the font, translating and explaining words and many more are available to the user. If you wish, you can test the performance of the Immersive Reader service on this page.
According to Microsoft’s promise, we can expect that Persian speech support will reach more tools and services from Microsoft. As mentioned, one of these tools is the desktop version and mobile platforms of the Office suite, and the addition of the ability to convert text to Persian speech to Word from the Office suite, which is one of the most widely used tools used by Iranian users, can greatly transform users. Create from this software.
How do you, the users of Zomit, evaluate the performance of the Read Aloud tool? Share your experiences in the comments section.