Research on a Mongolian Text to Speech Model Based on Ghost and ILPCnet
The core challenge of speech synthesis technology is how to convert text information into an audible audio form to meet the needs of users. In recent years, the quality of speech synthesis based on end-to-end speech synthesis models has been significantly improved. However, due to the characteristic...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-01-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/14/2/625 |
| _version_ | 1797340133925584896 |
|---|---|
| author | Qing-Dao-Er-Ji Ren Lele Wang Wenjing Zhang Leixiao Li |
| author_facet | Qing-Dao-Er-Ji Ren Lele Wang Wenjing Zhang Leixiao Li |
| author_sort | Qing-Dao-Er-Ji Ren |
| collection | DOAJ |
| description | The core challenge of speech synthesis technology is how to convert text information into an audible audio form to meet the needs of users. In recent years, the quality of speech synthesis based on end-to-end speech synthesis models has been significantly improved. However, due to the characteristics of the Mongolian language and the lack of an audio corpus, the Mongolian speech synthesis model has achieved few results, and there are still some problems with the performance and synthesis quality. First, the phoneme information of Mongolian was further improved and a Bang-based pre-training model was constructed to reduce the error rate of Mongolian phonetic synthesized words. Second, a Mongolian speech synthesis model based on Ghost and ILPCnet was proposed, named the Ghost-ILPCnet model, which was improved based on the Para-WaveNet acoustic model, replacing ordinary convolution blocks with stacked Ghost modules to generate Mongolian acoustic features in parallel and improve the speed of speech generation. At the same time, the improved vocoder ILPCnet had a high synthesis quality and low complexity compared to other vocoders. Finally, a large number of data experiments were conducted on the proposed model to verify its effectiveness. The experimental results show that the Ghost-ILPCnet model has a simple structure, fewer model generation parameters, fewer hardware requirements, and can be trained in parallel. The average subjective opinion score of its synthesized speech reached 4.48 and the real-time rate reached 0.0041. It ensures the naturalness and clarity of synthesized speech, speeds up the synthesis speed, and effectively improves the performance of the Mongolian speech synthesis model. |
| first_indexed | 2024-03-08T09:58:34Z |
| format | Article |
| id | doaj.art-30ab694428834de7b9bd8c306c23b534 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-08T09:58:34Z |
| publishDate | 2024-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-30ab694428834de7b9bd8c306c23b5342024-01-29T13:43:12ZengMDPI AGApplied Sciences2076-34172024-01-0114262510.3390/app14020625Research on a Mongolian Text to Speech Model Based on Ghost and ILPCnetQing-Dao-Er-Ji Ren0Lele Wang1Wenjing Zhang2Leixiao Li3School of Information Engineering, Inner Mongolia University of Technology, Hohhot 010051, ChinaSchool of Information Engineering, Inner Mongolia University of Technology, Hohhot 010051, ChinaSchool of Information Engineering, Inner Mongolia University of Technology, Hohhot 010051, ChinaCollege of Data Science and Application, Inner Mongolia University of Technology, Hohhot 010051, ChinaThe core challenge of speech synthesis technology is how to convert text information into an audible audio form to meet the needs of users. In recent years, the quality of speech synthesis based on end-to-end speech synthesis models has been significantly improved. However, due to the characteristics of the Mongolian language and the lack of an audio corpus, the Mongolian speech synthesis model has achieved few results, and there are still some problems with the performance and synthesis quality. First, the phoneme information of Mongolian was further improved and a Bang-based pre-training model was constructed to reduce the error rate of Mongolian phonetic synthesized words. Second, a Mongolian speech synthesis model based on Ghost and ILPCnet was proposed, named the Ghost-ILPCnet model, which was improved based on the Para-WaveNet acoustic model, replacing ordinary convolution blocks with stacked Ghost modules to generate Mongolian acoustic features in parallel and improve the speed of speech generation. At the same time, the improved vocoder ILPCnet had a high synthesis quality and low complexity compared to other vocoders. Finally, a large number of data experiments were conducted on the proposed model to verify its effectiveness. The experimental results show that the Ghost-ILPCnet model has a simple structure, fewer model generation parameters, fewer hardware requirements, and can be trained in parallel. The average subjective opinion score of its synthesized speech reached 4.48 and the real-time rate reached 0.0041. It ensures the naturalness and clarity of synthesized speech, speeds up the synthesis speed, and effectively improves the performance of the Mongolian speech synthesis model.https://www.mdpi.com/2076-3417/14/2/625Mongolian speech synthesisnon-autoregressiveGhost modulevocoder |
| spellingShingle | Qing-Dao-Er-Ji Ren Lele Wang Wenjing Zhang Leixiao Li Research on a Mongolian Text to Speech Model Based on Ghost and ILPCnet Applied Sciences Mongolian speech synthesis non-autoregressive Ghost module vocoder |
| title | Research on a Mongolian Text to Speech Model Based on Ghost and ILPCnet |
| title_full | Research on a Mongolian Text to Speech Model Based on Ghost and ILPCnet |
| title_fullStr | Research on a Mongolian Text to Speech Model Based on Ghost and ILPCnet |
| title_full_unstemmed | Research on a Mongolian Text to Speech Model Based on Ghost and ILPCnet |
| title_short | Research on a Mongolian Text to Speech Model Based on Ghost and ILPCnet |
| title_sort | research on a mongolian text to speech model based on ghost and ilpcnet |
| topic | Mongolian speech synthesis non-autoregressive Ghost module vocoder |
| url | https://www.mdpi.com/2076-3417/14/2/625 |
| work_keys_str_mv | AT qingdaoerjiren researchonamongoliantexttospeechmodelbasedonghostandilpcnet AT lelewang researchonamongoliantexttospeechmodelbasedonghostandilpcnet AT wenjingzhang researchonamongoliantexttospeechmodelbasedonghostandilpcnet AT leixiaoli researchonamongoliantexttospeechmodelbasedonghostandilpcnet |