ASR read-aloud compare-and-repeat self-study strategy for pronunciation practice¶

Strategy · All strategies
Evidence · no claims cited

Description¶

A self-study routine in which learners select a sentence from a set passage, activate ASR (e.g., the microphone icon in Google Documents or the Google Translate and Papago keyboards), speak the sentence, and compare the transcribed output to the target sentence. As the article describes it, students "compare the text output on the smartphone to the selected sentence" and repeat until there is a perfect match, treating mismatches as evidence of specific pronunciation errors. The treatment group practiced this way 15 minutes per week in class for four weeks and was encouraged to continue at home.

Design Implications¶

Context¶

Requirements¶

• Access to ASR voice typing on smartphones or Google Documents
• Training in ASR's features and limitations, including that ASR works better with full sentences than isolated words, is sensitive to speaking rate and pauses, and is affected by noise

Constraints¶

• ASR does not record suprasegmentals (stress, rhythm, intonation), which the article notes critics consider critical to speaking proficiency

Target Learners¶

• Korean university EFL students
• tertiary L2 English learners

Target Learning Goals¶

• segmental pronunciation accuracy
• self-directed pronunciation practice

Related Strategies¶

• Dictation/Speech-to-text

Examples¶

-

Key Sources¶

• Dillon, Thomas, & Wells, Donald. (2023). Effects of pronunciation training using automatic speech recognition on pronunciation accuracy of Korean English language learners. English Teaching, 78(1), 3-23. https://doi.org/10.15858/engtea.78.1.202303.3