Pronunciation

Self-Assessment of L2 Pronunciation: Using Automatic Speech Recognition (ASR)

This study looks at self-assessment for evaluating second or foreign language (L2) competencies, more specifically pronunciation. It does so by examining whether automatic speech recognition (ASR) can help learners form a more accurate portrayal of their pronunciation abilities. At present, there are very few studies on ASR systems as a tool for refining learners' ability to better assess their own pronunciation skills in a second or foreign language.

For the study, you will participate in a short one-time session. You will complete a self-assessment of your pronunciation skills (10-15 minutes). Then you will partake training on how to use automatic speech recognition to read sentences aloud (about 15 minutes) followed by reading similar sentences out loud in the ASR. You will read more practice sentences before reading new sentences that will be assessed by evaluators (15 minutes). Finally, you will complete the self-assessment of your pronunciation again.

Prénom

Nom

Email

Adresse

Ville

Province

Code Postale

Téléphone

Date de Naissance

J’ai lu les informations sur la protection des données et je suis d’accord pour passer le test de classement en anglais (ou autre langue).

1.

Consent to save data

This study saves the test results (recordings, texts, and answers to test questions) in the database and are used for internal file management purposes. These data are confidential and secure.

I have read the information on data protection, and I agree to take the self-assessment test. I accept:

Yes No

1 out of 13

2.

Consent to use data for research

Study Title: Improving Self-Assessment of L2 Pronunciation: Leveraging Automatic Speech Recognition (ASR).

Researcher: Carey Nelson (PhDs student in Applied Linguistics)

Researcher’s Contact Information: carey.nelson@concordia.ca

Faculty Supervisors: Dr. Walcir Cardoso Applied Linguistics/Education, Dr. Heike Neumann, Applied Linguistics/Education

Faculty Supervisor’s Contact Information:

Dr. Walcir Cardoso: walcir.cardoso@concordia.ca; 514-848-2424 x2451;

Dr. Heike Neumann: heike.neumann@concordia.ca: 514-848-2424 x2443:

Source of funding for the study: None

____________________________________________________________________

You are being invited to participate in the research study mentioned above. This form provides information about what participating would mean. Please read it carefully before deciding if you want to participate or not. If there is anything you do not understand, or if you want more information, please ask the researcher.

PURPOSE: The purpose of the research is to investigate English Second Language user ability to self-assess pronunciation while using automatic speech recognition (ASR).
PROCEDURES: If you participate, you will be asked to complete the following tasks remotely (times are approximate):

Complete a questionnaire on self-assessment of pronunciation skills (15 min), perform an practice ASR dictation task of seven sentences followed by reading aloud seven sentences (15 min), which will be recorded (5 min), read aloud seven other sentences for practice purposes, followed by reading aloud seven sentences (15 min), which will be recorded (15 min), and filling out the questionnaire on self-assessment again.

The practice and read aloud tasks will be recorded (audio only).

In total, participating in this study will involve over an hour of completing the tasks online for 1-2 hours in total.

RISKS AND BENEFITS:There are no known risks associated with participating in this research. One benefit could be that you will learn about a tool that could help your ability to self-assess your pronunciation. As well, I am a trained ESL teacher and can provide feedback on your pronunciation if you wish.

CONFIDENTIALITY: We will gather the following information as part of this research:

We will collect your anonymous demographic information as well as self-assessment questionnaires. We will also collect the audio of you reading sentences aloud.
We will not allow anyone to access the information, except people directly involved in the research. We will only use the information for the purposes of the research described in this form.
The information gathered will be coded and interviews will be transcribed. It is not necessary to give your name, which means there will not be any reference to you anywhere in the written study. No-one else will have access to your information.
We will protect the information by keeping all digital files on a password protected computer.
We intend to publish the results of the research. However, it will not be possible to identify you in the published results.
We will destroy the information three years after the end of the study.

CONDITIONS OF PARTICIPATION

You do not have to participate in this research. It is purely your decision. If you do participate, you can stop at any time. You can also ask that the information you provided not be used, and your choice will be respected. If you decide that you don’t want us to use your information, you must tell the researcher before February 28, 2025.
As a compensatory indemnity for participating in this research, you will be entered into a draw for one of five gift cards of $20. If you withdraw before the end of the research, your name will remain in the draw.
There are no negative consequences for not participating, stopping in the middle, or asking us not to use your information.

PARTICIPANT’S DECLARATION: I have read and understood this form. I have had the chance to ask questions and any questions have been answered. I agree to participate in this research under the conditions described.

If you have questions about the scientific or scholarly aspects of this research, please contact the researcher. Their contact information is at the beginning. You may also contact their faculty supervisor (Dr. Walcir Cardoso: walcir.cardoso@concoria.ca) or Dr. Heike Neumann (heike.neumann@concordia.ca).

If you have concerns about ethical issues in this research, please contact the Manager, Research Ethics, Concordia University, 514.848.2424 ex. 7481 or oor.ethics@concordia.ca.

I have read the information and I accept that my anonymous test responses may be used for research purposes

Yes No

2 out of 13

3.

What is your age? (Choose the year)

3 out of 13

4.

What is your first language?

Portuguese

French

English

5.

4 out of 13

6.

I can easily and accurately guess what the sound is in an unknown word based on my existing knowledge.

Never

Always

5 out of 13

7.

I can self-correct if I mispronounce a word.

Never

Always

6 out of 13

8.

I can carry out a conversation without the listener asking me to repeat what I said.

Never

Always

7 out of 13

9.

I can carry out a conversation without mispronouncing words.

Never

Always

8 out of 13

10.

I can use English word stress, sentence stress, rhythm, and intonation to highlight the meaning of my message.

Never

Always

9 out of 13

11.

I can use English word stress, sentence stress, rhythm, and intonation with very little influence from the other language(s) that I speak

Never

Always

10 out of 13

12.

I can be completely intelligible to a listener regardless of any influences from the other language(s) I speak on my accent or the prosody I use.

Never

Always

11 out of 13

13.

To what extent do you sound like a highly proficient English speaker when you speak English?

Never

Always

12 out of 13

14.

How easy is it for a listener to understand your spoken English?

Never

Always

13 out of 13