We introduce a new collection of spoken English audio suitable for training speech recognition systems under limited or no supervision. It is derived from open-source audio books from the LibriVox project. It contains over 60K hours of audio, which is, to our knowledge, the largest freely-available corpus of speech. The audio has been segmented using voice activity detection and is tagged with SNR, speaker ID and genre descriptions. Additionally, we provide baseline systems and evaluation metrics working under three settings: (1) the zero resource/unsupervised setting (ABX), (2) the semisupervised setting (PER, CER) and (3) the distant supervision setting (WER). Settings (2) and (3) use limited textual resources (10 minutes to 10 hours) aligned with the speech. Setting (3) uses large amounts of unaligned text. They are evaluated on the standard LibriSpeech dev and test sets for comparison with the supervised state-of-the-art.Index Terms-unsupervised and semi-supervised learning, distant supervision, dataset, zero-and low resource ASR.

There is an implicit assumption that traditional hybrid approaches for automatic speech recognition (ASR) cannot directly model graphemes and need to rely on phonetic lexicons to get competitive performance, especially on English which has poor grapheme-phoneme correspondence. In this work, we show for the first time that, on English, hybrid ASR systems can in fact model graphemes effectively by leveraging tied context-dependent graphemes, i.e., chenones. Our chenone-based systems significantly outperform equivalent senone baselines by 4.5% to 11.1% relative on three different English datasets. Our results on Librispeech are state-ofthe-art compared to other hybrid approaches and competitive with previously published end-to-end numbers. Further analysis shows that chenones can better utilize powerful acoustic models and large training data, and require context-and position-dependent modeling to work well. Chenone-based systems also outperform senone baselines on proper noun and rare word recognition, an area where the latter is traditionally thought to have an advantage. Our work provides an alternative for end-to-end ASR and establishes that hybrid systems can be improved by dropping the reliance on phonetic knowledge.

In the petroleum industry, high temperature and high pressure (HTHP) would dramatically worsen rheological properties and increase fluid loss volumes of drilling fluids. Synthetic polymer as an indispensable additive has attracted more and more attention recently. In this article, a new copolymer (named AADS) of 2‐acrylamide‐2‐methylpropanesulfonic acid, acrylamide, dimethyl diallyl ammonium chloride, and sodium styrene sulfonate was synthesized through aqueous solution polymerization. The chemical structure of the copolymer was characterized by Fourier transform infrared spectroscopy and nuclear magnetic resonance. Moreover, its thermal stability was simultaneously analyzed using a differential scanning calorimetry. The results showed that the synthetic polymer contained all the designed functional groups, and its structure was consistent to the desired one. Under contamination of sodium chloride, AADS solution maintained relatively high viscosity in high concentration brine, showing a good antisalt capacity. Furthermore, the effect of AADS content and temperature on rheological behavior and fluid loss volume of the water‐based drilling fluid (WBDF) containing the synthesized product were investigated according to the American Petroleum Institute standard. Results showed that the rheological and filtration properties of the prepared WBDF were improved with the increase in the AADS concentration before and after the thermal aging test. In addition, in the temperature range of 80–240 °C, a reversible rheological behavior was observed during the heating–cooling process, and the HTHP fluid loss was controlled within 22.5 mL, suggesting that the copolymer AADS was suitable for making WBDF s with high temperature resistance. © 2019 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2019, 136, 47813.

No abstract