Automatic speech recognition for low-resource languages
Speech recognition technology has been used for a long time, but recognizing a speech accurately is a very difficult task. In this topic, we mainly use the conformer-ctc model provided by open-source toolkits (Nemo), and fine-tune the model to achieve better training results. If you are interested in automatic speech recognition, and have a good foundation in python, it is highly recommended that you choose this topic.
Automatic speech recognition for low-resource languages
Speech recognition technology has been used for a long time, but recognizing a speech accurately is a very difficult task. In this topic, we mainly use the conformer-ctc model provided by open-sourch toolkits(Nemo), and fine-tune the model to achieve better training results. If you are interested in automatic speech recognition, and have a good foundation in python, it is highly recommended that you choose this topic.
Automatic speech recognition for low-resource languages
Speech recognition technology has been used for a long time, but recognizing a speech accurately is a very difficult task. In this topic, we mainly use the conformer-ctc model provided by open-source toolkits (Nemo), and fine-tune the model to achieve better training results. If you are interested in automatic speech recognition, and have a good foundation in python, it is highly recommended that you choose this topic.
Automatic speech recognition for low-resource languages
Speech recognition technology has been used for a long time, but recognizing a speech accurately is a very difficult task. In this topic, we mainly use the conformer-ctc model provided by open-source toolkits (Nemo), and fine-tune the model to achieve better training results. If you are interested in automatic speech recognition, and have a good foundation in python, it is highly recommended that you choose this topic.
Automatic speech recognition for low-resource languages
Speech recognition technology has been used for a long time, but recognizing a speech accurately is a very difficult task. In this topic, we mainly use the conformer-ctc model provided by open-source toolkits (Nemo), and fine-tune the model to achieve better training results. If you are interested in automatic speech recognition, and have a good foundation in python, it is highly recommended that you choose this topic.
Automatic speech recognition for less-resource language
Focused on Automatic Speech Recognition (ASR)
1.Attempting to explore the implications of data augmentation solution for limited-resource languages.
2.trying to find the optimal data augmentation solution by using Automatic Data Augmentation technical.
Automatic speech recognition for less-resource language
Focused on Automatic Speech Recognition (ASR)
1.Attempting to explore the implications of data augmentation solution for limited-resource languages.
2.trying to find the optimal data augmentation solution by using Automatic Data Augmentation technical.