This study aims to investigate the acquisition of vocabulary recognition and vocabulary production in the short- and long-term via listening and reading comprehension activities using Voice® software. Sixty participants were invited to read or listen to two passages in different sessions, that is, three sessions in listening inputs and one session in reading comprehension. The materials in reading comprehension input were converted into audio and given for the students who were assigned to participate in the listening sessions. Vocabulary recognition was measured through written multiple-choice tests, whereas the vocabulary production was assessed by a cloze test for both scenarios. Each test was administered before the intervention immediately after the intervention and at a delayed post of time for both inputs. Results showed that the students were able to learn new vocabulary from both modes in the pre-post-delayed treatments. Participants scored higher in recognition vocabulary test than in the production test. In view of the vocabulary gains in both scenarios in the post-tests, the performance of listeners was significantly lower than that of readers after the first session, which was almost the same after the second session, and significantly higher after the third listening session. However, vocabulary acquisition from listening comprehension mode was significantly higher than that from reading input in the delayed post-tests.