Making an English Speech Resemble the User’s Voice Using UTAU and Interactive Evolutionary Computation

Miyamoto, Taichi; Gan, Haoran; FUKUMOTO, Makoto

doi:10.5057/isase.2022-c000024

Cited by 3 publications

(9 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides, we would like to mention that this method is so versatile since we are voice-changing the PC's sounds. In our previous research [2], the biggest issue was the lack of real-time response because the method could apply effectors only on recorded voices. However, in the second experiment, we asked ChatGPT to make some English sentences that are understandable for users, and the 2 types of voices read those sentences.…”

Section: Discussionmentioning

confidence: 99%

“…Unfortunately, Japanese people struggle while speaking English because of the differences between those languages. We have been trying to research a method that can help English learners imitate the pronunciation of native speakers, such as making English voices similar to users' voices [1] [2]. The purpose of this study is the same, however, we try to reach the goal from another point of view.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Making English Voices Similar to User’s Voices Using Voice Changer and Interactive Differential Evolution

MIYAMOTO,

FUKUMOTO

2024

ISASE

View full text Add to dashboard Cite

English skills are important for college students when they submit papers to international conferences. We propose a method to practice speaking skills by using a voice changer and Interactive Differential Evolution, which makes English voices similar to users' voices inside of their heads. This method should be useful when users try to imitate any English voices that are different from the users. 11 examinees participated in the experiments through 10 generations with 6 vectors of Differential Evolution in each generation. They kept choosing the better option in terms of the similarity of the voices compared to the examinees' voices. Unfortunately, the transition of the parameters was not significant, and we could not observe any differences between the initial and final generations. However, we also conducted another evaluation experiment, and the results showed that the system could successfully make English voices similar to users' voices inside of their heads. Besides, compared to our previous method, this method is so versatile that we can use any sound resources as a factor of the experiment since this system is voice-changing the PC's output sound. In the future, we would reconsider the number of the parameters and check the effect of them.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Making English Voices Similar to User’s Voices Using Voice Changer and Interactive Differential Evolution

MIYAMOTO,

FUKUMOTO

2024

ISASE

View full text Add to dashboard Cite

show abstract

“…Interactive Evolutionary Computation (IEC) was known as a method that obtains the user's media content based on a stochastic search [6,7], and previous studies created sound content using IECs [8][9][10][11][12][13][14]. The purpose of this study is to propose a method to produce the voice data of the model pronunciation similar to the user's voice by using UTAU [15] and interactive differential evolution (IDE) [16], one of the IECs.…”

Section: Introductionmentioning

confidence: 99%

“…In a previous study, we constructed a similar system by using UTAU and an interactive genetic algorithm (IGA), which is one of the IEC [14]. In the experiment, the result of reevaluation comparing the initial and final generation was not good, and some examinees felt fatigued because of the long experiment duration.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Making an English Speech Similar to the User’s Voice using UTAU and Interactive Differential Evolution

Miyamoto

Gan

FUKUMOTO

2023

IJAE

Self Cite

View full text Add to dashboard Cite

Practicing English pronunciation is difficult for non-native speakers because of the differences in vowels and consonants. There are several ways to practice them such as Shadowing, however, if the voice's features greatly differ from the learner's voice, it should be difficult for learners to reproduce. To solve this problem, we propose a method to make the pronunciation data of the model pronunciation similar to the learner's voice by using UTAU and Interactive Differential Evolution. A listening experiment was conducted with the concrete system of IDE and UTAU. Twelve examinees participated in the experiment through ten generations based on paired comparisons for making the voices similar to their own voices inside their heads. As a result, we could successfully make the voices similar to the examinees' voices. Since it has paired comparison, we believe that the paired comparison-based IDE is a better method than the general Interactive Genetic Algorithm with scoring.

show abstract

Making English Voices Similar to the User’s Voice by Using Interactive Differential Evolution

Miyamoto

FUKUMOTO

2023

ISASE

Self Cite

View full text Add to dashboard Cite

English skills are becoming necessary for students who would like to give a presentation at international conferences. However, it is very difficult for non-native English speakers to speak in English because of the differences between English and their native languages. There are some ways to practice speaking English such as Shadowing, and if the audio is similar to the learner's voices inside their heads, it should be more comfortable for the learners to listen to it and repeat after the audio. We propose a method to make English voices similar to the user's voice by using Interactive Differential Evolution. Experiments were conducted with the concrete system of IDE. 9 examinees participated in the experiments through 10 generations with 6 vectors in each generation and kept choosing the better option. After the experiment, we randomly chose 2 voices from the initial and final generation and asked the examinees to evaluate them. As a result, significantly higher evaluations were obtained in the voices from the final generation.

show abstract

Making an English Speech Resemble the User’s Voice Using UTAU and Interactive Evolutionary Computation

Cited by 3 publications

References 2 publications

Making English Voices Similar to User’s Voices Using Voice Changer and Interactive Differential Evolution

Making English Voices Similar to User’s Voices Using Voice Changer and Interactive Differential Evolution

Making an English Speech Similar to the User’s Voice using UTAU and Interactive Differential Evolution

Making English Voices Similar to the User’s Voice by Using Interactive Differential Evolution

Contact Info

Product

Resources

About