Automated speech therapy through personalized pronunciation correction using reinforcement learning and large language models
Traditional approaches to pronunciation correction often face challenges in personalization, adaptability, and consistent feedback. This study introduces a novel AI-powered system that integrates Reinforcement Learning (RL) and Large Language Models (LLMs) to address these limitations. The system em...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-03-01
|
| Series: | Results in Engineering |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2590123025000313 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841555671881875456 |
|---|---|
| author | Ritika Lakshminarayanan Ayesha Shaik Ananthakrishnan Balasundaram |
| author_facet | Ritika Lakshminarayanan Ayesha Shaik Ananthakrishnan Balasundaram |
| author_sort | Ritika Lakshminarayanan |
| collection | DOAJ |
| description | Traditional approaches to pronunciation correction often face challenges in personalization, adaptability, and consistent feedback. This study introduces a novel AI-powered system that integrates Reinforcement Learning (RL) and Large Language Models (LLMs) to address these limitations. The system employs a custom Proximal Policy Optimization (PPO) algorithm for precise pronunciation evaluation and an Large Language Models to deliver detailed, encouraging, and user-specific feedback. It was evaluated using the CMU Sphinx Dictionary dataset, a foundational phonetic resource, alongside dynamically generated user-specific session data for personalized feedback and model refinement. Further validation utilized datasets such as TIMIT, LibriTTS, SpeechOcean762, and the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS), enabling direct comparisons with contemporary methods. Results demonstrate the system's robustness in handling diverse phonetic variations. While primarily tested on English data, its modular architecture supports adaptation to other languages and dialects through language-specific phonetic datasets. The system achieved exceptional performance metrics: 97.9 % phoneme-level accuracy, 87.7 % word-level accuracy, 95.2 % syllable count accuracy, and 89.4 % perfect accuracy on the CMU Sphinx dataset. This innovative approach underscores the potential of advanced AI techniques to enhance the personalization and effectiveness of pronunciation correction systems. All findings are quantitatively validated and thoroughly documented. |
| format | Article |
| id | doaj-art-4d3c5a63f38149059168a1e6391097b4 |
| institution | Kabale University |
| issn | 2590-1230 |
| language | English |
| publishDate | 2025-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Results in Engineering |
| spelling | doaj-art-4d3c5a63f38149059168a1e6391097b42025-01-08T04:53:25ZengElsevierResults in Engineering2590-12302025-03-0125103943Automated speech therapy through personalized pronunciation correction using reinforcement learning and large language modelsRitika Lakshminarayanan0Ayesha Shaik1Ananthakrishnan Balasundaram2School of Computer Science and Engineering, Vellore Institute of Technology, Chennai 600127, IndiaCentre for Cyber Physical Systems, Vellore Institute of Technology, Chennai 600127, India; Corresponding author.Centre for Cyber Physical Systems, Vellore Institute of Technology, Chennai 600127, IndiaTraditional approaches to pronunciation correction often face challenges in personalization, adaptability, and consistent feedback. This study introduces a novel AI-powered system that integrates Reinforcement Learning (RL) and Large Language Models (LLMs) to address these limitations. The system employs a custom Proximal Policy Optimization (PPO) algorithm for precise pronunciation evaluation and an Large Language Models to deliver detailed, encouraging, and user-specific feedback. It was evaluated using the CMU Sphinx Dictionary dataset, a foundational phonetic resource, alongside dynamically generated user-specific session data for personalized feedback and model refinement. Further validation utilized datasets such as TIMIT, LibriTTS, SpeechOcean762, and the Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS), enabling direct comparisons with contemporary methods. Results demonstrate the system's robustness in handling diverse phonetic variations. While primarily tested on English data, its modular architecture supports adaptation to other languages and dialects through language-specific phonetic datasets. The system achieved exceptional performance metrics: 97.9 % phoneme-level accuracy, 87.7 % word-level accuracy, 95.2 % syllable count accuracy, and 89.4 % perfect accuracy on the CMU Sphinx dataset. This innovative approach underscores the potential of advanced AI techniques to enhance the personalization and effectiveness of pronunciation correction systems. All findings are quantitatively validated and thoroughly documented.http://www.sciencedirect.com/science/article/pii/S2590123025000313Automatic speech recognitionReinforcement learningProximal policy optimizationLarge language modelPhonetic transcriptionSpeech synthesis markup language |
| spellingShingle | Ritika Lakshminarayanan Ayesha Shaik Ananthakrishnan Balasundaram Automated speech therapy through personalized pronunciation correction using reinforcement learning and large language models Results in Engineering Automatic speech recognition Reinforcement learning Proximal policy optimization Large language model Phonetic transcription Speech synthesis markup language |
| title | Automated speech therapy through personalized pronunciation correction using reinforcement learning and large language models |
| title_full | Automated speech therapy through personalized pronunciation correction using reinforcement learning and large language models |
| title_fullStr | Automated speech therapy through personalized pronunciation correction using reinforcement learning and large language models |
| title_full_unstemmed | Automated speech therapy through personalized pronunciation correction using reinforcement learning and large language models |
| title_short | Automated speech therapy through personalized pronunciation correction using reinforcement learning and large language models |
| title_sort | automated speech therapy through personalized pronunciation correction using reinforcement learning and large language models |
| topic | Automatic speech recognition Reinforcement learning Proximal policy optimization Large language model Phonetic transcription Speech synthesis markup language |
| url | http://www.sciencedirect.com/science/article/pii/S2590123025000313 |
| work_keys_str_mv | AT ritikalakshminarayanan automatedspeechtherapythroughpersonalizedpronunciationcorrectionusingreinforcementlearningandlargelanguagemodels AT ayeshashaik automatedspeechtherapythroughpersonalizedpronunciationcorrectionusingreinforcementlearningandlargelanguagemodels AT ananthakrishnanbalasundaram automatedspeechtherapythroughpersonalizedpronunciationcorrectionusingreinforcementlearningandlargelanguagemodels |