Towards Reading Aloud Instruction Using Model Voices

（This is an English version of my previous post, モデル音声を使った音読指導に向けて.）

When practicing reading English texts aloud, it is essential to work with a model voice. However, no matter how much we tell them about the importance of audio materials, some students seem to fail to incorporate model voices into their own learning.

In this article, I will consider why high school students are so reluctant to reach out to model audio, and offer some suggestions of my own.

Reluctance to play audio materials

The first point is obvious; learners, especially ordinary junior and senior high school students, find it troublesome to take out a CD and play the model audio when they want to start reading aloud in their home study. In recent years, there has been an increase in materials that use QR codes to access audio from smartphones instead of CDs, but even this shortcut is still seen as tiresome.

Basically, the only way is to have students understand and realize the importance and effectiveness of reading aloud using audio materials, and make them willing to take the time to do so. It will be easier to read aloud using model voices when digital textbooks become more widely available for each student and model voices can be played from the textbook (tablet) in front of the student with a single touch.

Model voices being too fast

Often times, even if a model voice is played on a CD or accessed from a QR code, the voice is too fast to be suitable for reading aloud practice. Most model voices are read out too fluently and without hesitation, and this speed is not suitable for practicing reading aloud, especially for "storytelling-like" practice.

Of course, the model voice for such fluent reading is intended to allow the voice to be used for listening activities as well. It may also indicate the standard of the speed at which the reading should ultimately be comprehended when it is read silently. Certainly, there is a need and demand for such model voices. But, in addition to such fast-read model audio, would it be possible to prepare a bit slower model voice suitable for storytelling-like reading aloud? It would be very useful not only for teachers but also for learners themselves if you could prepare a model audio for reading-aloud mode in addition to the normal speed version.

Many textbooks and teaching materials have audio with pauses at breaks in the text (like slashes). This is very useful for activities such as repetition and sight translation, but it would be of great help if we could have a "storytelling" mode as well. In the storytelling mode, not only should the speed be slow, but the pauses, tone, and intonation should be emphasized as much as possible to make it sound as if you are "storytelling" to the students.

These days, software and applications that read English sentences aloud to you have evolved considerably. Until recently, they were monotone machine voices, but they are now capable of reading out loud at a level comparable to that of a human reader. If the technology continues to evolve at this rate, we may be able to create a "read-aloud mode" within the speechreading function. We hope to see such a function in the near future that allows the AI to determine the parts of the text that need emphasis and the appropriate pauses, and to read the text with more emphasis on the pitch of the words and the intonation.

Use more speech materials

What kind of subject matter and text genres are best suited for reading aloud in the storytelling mode? I think speech materials are probably the most appropriate. Speech material has many elements that you want the students to be aware of when they read aloud, such as tone, intonation, appropriate pauses, gestures, etc. It would be easier to read aloud with more emotion if you could use the speaker's own voice as the material. For instance, while the difficulty level might be an issue, Steve Jobs' and Severn Suzuki's speeches are well-known and are often used in English textbooks, and there are many great speeches on TED Talks that we want to expose our students to. By practicing reading these speeches aloud using their real voices as models, I think students will be able to work on areas that are difficult to focus on when reading editorial or expository texts aloud.

In reality, from various perspectives, it may be difficult to treat these authentic speeches as they are as teaching materials. However, I am sure that these speech materials are suitable for reading aloud instruction with an eye toward storytelling-like reading aloud. It would be nice to use materials that imitate speeches as a bridge from short and conversational English reading practice in junior high school to longer reading practice in high school English.

In this article, I have considered how model speech and teaching materials can be used to help students learn to read aloud better. We hope this will be of some help in selecting and creating teaching materials.