Feature bandwidth extension for Persian conversational telephone speech recognition

Goodarzi, Mohammad Mohsen; Almasganj, Farshad; Kabudian, Jahanshah; Shekofteh, Yasser; Rezaei, Iman Sarraf

doi:10.1109/iraniancee.2012.6292541

Cited by 5 publications

(9 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…, L (3) e1 eL( 2) eL(2)+1 e2L (2) eL( 3) In an SRNN, the generation of each output sample depends on all previous and current input samples. However, this plain LSTM-RNN architecture still has some deficiencies for waveform modeling and generation.…”

Section: A Sample-level Recurrent Neural Networkmentioning

confidence: 99%

“…t denotes the t-th waveform frame at the k-th tier, and c (k) is the number of concatenated frames at the k-th tier. We have c (3) = c (2) = 1 in the model structure shown in Fig. 3.…”

Section: B Hierarchical Recurrent Neural Networkmentioning

confidence: 99%

“…Some frame-level auxiliary features extracted from input narrowband waveforms, such as bottleneck (BN) features [36], have shown their effectiveness in improving the performance , L (3) L (4) -L (3) +1, , L (4) y1 yL(2) of vocoder-based BWE [32]. In order to combine such auxiliary inputs with the HRNN model introduced in Section III-B, a conditional HRNN structure is designed as shown in Fig.…”

Section: Conditional Hierarchical Recurrent Neural Networkmentioning

confidence: 99%

“…Therefore, there were two LSTM layers and two FF layers in total which was the same as the SRNN system. The number of c (k) , k = {1, 2, 3} in ( 14) and ( 19) were set as c (3) = c (2) = 2, c (1) = L (2) in our experiments after tuning on the validation set. Some other setups, such as the dimension of the hidden units and the training method, were the same as that of the SRNN system mentioned above.…”

Section: A Experimental Setupmentioning

confidence: 99%

“…Therefore, speech bandwidth extension (BWE), which aims to restore the missing high-frequency components of narrowband speech using the correlations that exist between the low and high-frequency components of the wideband speech signal, has attracted the attentions of many researchers. BWE methods can not only be applied to real-time voice communication, but also benefit other speech signal processing areas such as text-tospeech (TTS) synthesis [1], speech recognition [2], [3], and speech enhancement [4], [5].…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Waveform Modeling and Generation Using Hierarchical Recurrent Neural Networks for Speech Bandwidth Extension

Ling

et al. 2018

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

This paper presents a waveform modeling and generation method using hierarchical recurrent neural networks (HRNN) for speech bandwidth extension (BWE). Different from conventional BWE methods which predict spectral parameters for reconstructing wideband speech waveforms, this BWE method models and predicts waveform samples directly without using vocoders. Inspired by SampleRNN which is an unconditional neural audio generator, the HRNN model represents the distribution of each wideband or high-frequency waveform sample conditioned on the input narrowband waveform samples using a neural network composed of long short-term memory (LSTM) layers and feed-forward (FF) layers. The LSTM layers form a hierarchical structure and each layer operates at a specific temporal resolution to efficiently capture long-span dependencies between temporal sequences. Furthermore, additional conditions, such as the bottleneck (BN) features derived from narrowband speech using a deep neural network (DNN)-based state classifier, are employed as auxiliary input to further improve the quality of generated wideband speech. The experimental results of comparing several waveform modeling methods show that the HRNN-based method can achieve better speech quality and runtime efficiency than the dilated convolutional neural network (DCNN)-based method and the plain sample-level recurrent neural network (SRNN)-based method. Our proposed method also outperforms the conventional vocoder-based BWE method using LSTM-RNNs in terms of the subjective quality of the reconstructed wideband speech.

show abstract

Section: A Sample-level Recurrent Neural Networkmentioning

confidence: 99%

“…t denotes the t-th waveform frame at the k-th tier, and c (k) is the number of concatenated frames at the k-th tier. We have c (3) = c (2) = 1 in the model structure shown in Fig. 3.…”

Section: B Hierarchical Recurrent Neural Networkmentioning

confidence: 99%