Robust Persian Digit Recognition in Noisy Environments Using Hybrid CNN-BiGRU Model
Fuente:
arXiv
Salvato in:
| Autori principali: | , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866912227936698368 |
|---|---|
| author | Nasr-Esfahani, Ali Bekrani, Mehdi Rajabi, Roozbeh |
| author_facet | Nasr-Esfahani, Ali Bekrani, Mehdi Rajabi, Roozbeh |
| contents | Artificial intelligence (AI) has significantly advanced speech recognition applications. However, many existing neural network-based methods struggle with noise, reducing accuracy in real-world environments. This study addresses isolated spoken Persian digit recognition (zero to nine) under noisy conditions, particularly for phonetically similar numbers. A hybrid model combining residual convolutional neural networks and bidirectional gated recurrent units (BiGRU) is proposed, utilizing word units instead of phoneme units for speaker-independent recognition. The FARSDIGIT1 dataset, augmented with various approaches, is processed using Mel-Frequency Cepstral Coefficients (MFCC) for feature extraction. Experimental results demonstrate the model's effectiveness, achieving 98.53%, 96.10%, and 95.92% accuracy on training, validation, and test sets, respectively. In noisy conditions, the proposed approach improves recognition by 26.88% over phoneme unit-based LSTM models and surpasses the Mel-scale Two Dimension Root Cepstrum Coefficients (MTDRCC) feature extraction technique along with MLP model (MTDRCC+MLP) by 7.61%. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_10857 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Robust Persian Digit Recognition in Noisy Environments Using Hybrid CNN-BiGRU Model Nasr-Esfahani, Ali Bekrani, Mehdi Rajabi, Roozbeh Sound Computer Vision and Pattern Recognition Audio and Speech Processing Artificial intelligence (AI) has significantly advanced speech recognition applications. However, many existing neural network-based methods struggle with noise, reducing accuracy in real-world environments. This study addresses isolated spoken Persian digit recognition (zero to nine) under noisy conditions, particularly for phonetically similar numbers. A hybrid model combining residual convolutional neural networks and bidirectional gated recurrent units (BiGRU) is proposed, utilizing word units instead of phoneme units for speaker-independent recognition. The FARSDIGIT1 dataset, augmented with various approaches, is processed using Mel-Frequency Cepstral Coefficients (MFCC) for feature extraction. Experimental results demonstrate the model's effectiveness, achieving 98.53%, 96.10%, and 95.92% accuracy on training, validation, and test sets, respectively. In noisy conditions, the proposed approach improves recognition by 26.88% over phoneme unit-based LSTM models and surpasses the Mel-scale Two Dimension Root Cepstrum Coefficients (MTDRCC) feature extraction technique along with MLP model (MTDRCC+MLP) by 7.61%. |
| title | Robust Persian Digit Recognition in Noisy Environments Using Hybrid CNN-BiGRU Model |
| topic | Sound Computer Vision and Pattern Recognition Audio and Speech Processing |
| url | https://arxiv.org/abs/2412.10857 |