Development and Implementation of a Voice Recognition Application for the Improvement and Feedback of English Pronunciation

The present study addresses the challenge of pronunciation in English learning, a crucial aspect for effective and fluent communication. Despite the plethora of online tools for English learning, there is a lack of specific solutions to enhance pronunciation, especially those that are freely accessible. In this context, a software was developed based on Python libraries that employ voice recognition technologies to analyze and provide real-time feedback on the user’s pronunciation. The system compares the user’s pronunciation with an original text and offers a quantitative accuracy metric. The software was tested with users of different linguistic skill levels (beginner, intermediate, and advanced), demonstrating its adaptability and effectiveness in pinpointing areas for improvement. Despite its limitations, such as not recognizing punctuation marks and interpreting long pauses as endings, the software shows significant potential for use in educational settings and online learning platforms. Furthermore, a notable progress was observed in the students' performance throughout the sessions, highlighting the software’s ability to promote learning and continuous improvement.