You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
47 lines
2.8 KiB
47 lines
2.8 KiB
From text to speech: The MITalk system
|
|
|
|
source for preparing the typed versions of the passages used in the reading com-
|
|
prehension condition. All fifteen passages were recorded on audio tape at a speak-
|
|
ing rate in excess of 180 words/minute for later playback. Two sets of response
|
|
booklets were prepared, one for the listening group and one for the reading group.
|
|
The booklets, which contained a varying number of multiple-choice questions
|
|
keyed to each paragraph, were arranged in order according to the presentation
|
|
schedule of the paragraphs on the audio tape. The booklets for subjects in the
|
|
reading group also included a typed copy of the passage immediately before the
|
|
appropriate set of questions. Appendix G also provides the set of questions cor-
|
|
responding to the passage.
|
|
|
|
13.4.1.3 Procedure Half of the forty-four subjects were assigned to the listening
|
|
group and the other half to the reading group. Subjects assigned to the reading
|
|
group were tested together in a classroom, while the subjects in the listening group
|
|
were tested in small groups of five or six subjects each using the listening facilities
|
|
of the previous studies. These subjects wore headphones and listened to the pas-
|
|
sages under the same conditions as the earlier subjects.
|
|
|
|
Instructions to the subjects in both groups emphasized that the purpose of the
|
|
study was to evaluate how well individuals could comprehend and understand con-
|
|
tinuous synthetic speech produced by a reading machine. Subjects in the listening
|
|
group were told that they would hear narrative passages about a wide variety of
|
|
topics and that their task was to answer the multiple-choice questions that were
|
|
keyed to the particular passages as best as they could based on the information
|
|
contained in the passages they heard. Similar instructions were provided to the
|
|
reading comprehension group.
|
|
|
|
As in the previous word-recognition study, the listening comprehension group
|
|
was presented with test passages in a self-paced format with the experimenter
|
|
present in the testing room operating the tape recorder via remote control. A given
|
|
test passage was presented only once for listening, after which, subjects im-
|
|
mediately turned their booklets to the appropriate set of test questions.
|
|
|
|
The subjects in the reading comprehension group were permitted to read each
|
|
passage only once and were explicitly told that they should not go back over the
|
|
passage after reading it or while answering the questions. This procedure was a
|
|
departure from the typical methods used in administering standardized reading
|
|
comprehension tests. Usually, the test passage is available to the subject for in-
|
|
spection and re-reading during the entire testing session. However, for present
|
|
purposes, we felt that comparisons between reading and listening comprehension
|
|
might be more closely matched by limiting exposure to one pass through the
|
|
materials.
|
|
|
|
164
|