Text-to-Speech technology converts written text into spoken words. In Speakable, it is primarily used in “Repeat” actions and media blocks added by teachers. Amazon Polly provides high-quality voices for many languages, while operating system voices are used when Amazon Polly voices are unavailable.

Speech-to-Text technology converts spoken words into written text. This functionality is used in “Repeat” actions to check pronunciation accuracy and in spoken responses for free response actions. Speakable leverages Google Chrome’s Web Speech API, which supports many languages for client-side applications.

Translation functionality provides translations of the target language text into the student’s native language. This feature is used in “Repeat” actions to display the translation after the student completes the action. Speakable utilizes Google Translate to support a vast array of languages for text translation.

Large Language Models are used for processing text and providing feedback. In Speakable, LLMs are applied in free response actions for auto-grading and feedback. OpenAI’s models support numerous languages, enabling comprehensive feedback and auto-grading capabilities.

The Speakable public activity library is a dynamic and growing collection of educational activities crafted by educators from around the globe. This library offers a diverse array of ready-to-use activities, designed to engage students and enhance their learning experiences. Educators can explore, share, and adapt these activities to suit their unique classroom needs, fostering a collaborative and innovative teaching community