Robust signal-to-noise ratio estimation based on waveform amplitude distribution analysis

By Chanwoo Kim, Richard M. Stern. 2008

227 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Yu Zhang
2. James R. Glass
3. Junichi Yamagishi
4. Wei-Ning Hsu
5. Ye Jia
6. Xu Tan
7. Yonghui Wu
8. Tao Qin

In Signal Processing

1. Yu Zhang
2. Junichi Yamagishi
3. James R. Glass
4. Wei-Ning Hsu
5. Yonghui Wu
6. Heiga Zen
7. Ye Jia
8. Ron J. Weiss

Citations in the corpus

• LibriTTS: A Corpus Derived from LibriSpeech for Text-to-Speech (2019, methodology): “Audio with WADA-SNR 20dB and 0dB are filtered out from the “clean” and “other” subsets, respectively.”
• Hierarchical Generative Modeling for Controllable Speech Synthesis (2018, methodology): “To quantify the noise control results, we used waveform amplitude distribution analysis (WADA) (Kim & Stern, 2008) to estimate an SNR without a reference clean signal.”
• Supervised and Unsupervised Speech Enhancement Using Nonnegative Matrix Factorization (2013, methodology): “Moreover, for the BNMF-based approaches the long-term SNR was estimated using [44] and we used Fig.”
• Voice Impersonation Using Generative Adversarial Networks (2018, methodology): “To evaluate the quality of our generated speech signal [18], we also conduct a signal-to-noise (SNR) ratio test using the standard NIST STNR method and the WADA SNR method [19].”
• Disentangling Correlated Speaker and Noise for Speech Synthesis via Data Augmentation and Adversarial Factorization (2019, methodology): “To quantify the ability to control noise, we use waveform amplitude distribution analysis (WADA) [28] to estimate an SNR without a clean reference signal.”
• Pretraining Strategies, Waveform Model Choice, and Acoustic Configurations for Multi-Speaker End-to-End Speech Synthesis (2020, methodology): “We also report estimated signal-tonoise ratio (SNR) based on the WADA SNR algorithm [25], as an objective measure of corpus quality.”
• An algorithm to improve speech recognition in noise for hearing-impaired listeners. (2013, unknown): “We consider SNR mismatch to be of less concern than noise mismatch because SNR estimation can be performed with reasonable accuracy (Kim and Stern, 2008; Narayanan and Wang, 2012).”