In many English as a foreign language (EFL) settings, classroom practice is often the only opportunity for oral development. Extensive reading (ER), self‑selected reading of large quantities of accessible texts, has consistently shown benefits for vocabulary and reading fluency, but its effect on student talking time (STT) and grammatical knowledge in low‑input contexts lacks rigorous experimental evidence. This randomized controlled trial examined a 14‑week ER program combined with structured speaking tasks among 90 intermediate male EFL learners (aged 14–18) at a private language institute in Alvand, Iran. Participants were randomly assigned to an ER + Speaking group (n = 30), a Speaking Practice Only group (n = 30), and a Control group (n = 30). All groups completed a pre‑test, an immediate post‑test, and 4‑week delayed post‑test measures of reading comprehension, speaking performance (fluency, accuracy, complexity, interaction, communicative effectiveness), grammatical knowledge (discrete‑point test, elicited imitation, spoken accuracy), and vocabulary size. The ER + Speaking group significantly outperformed both comparison groups on STT (d = 1.02–1.42) and grammar (d = 0.75–1.09), with gains largely maintained at follow‑up. Vocabulary growth partially mediated reading amount on STT, and each additional book read increased STT scores by approximately 0.35 points. The findings demonstrate that integrating ER with output practice enhances oral interaction and incidental grammar acquisition in input‑poor EFL classrooms, with reading volume playing a critical role.