P
Pedro J. Moreno
Researcher at Google
Publications - 128
Citations - 7944
Pedro J. Moreno is an academic researcher from Google. The author has contributed to research in topics: Language model & Word error rate. The author has an hindex of 45, co-authored 118 publications receiving 7206 citations. Previous affiliations of Pedro J. Moreno include Carnegie Mellon University & Hewlett-Packard.
Papers
More filters
Patent
Computer method and apparatus for uniform representation of genome sequences
TL;DR: A comparison database stores a predefined number of known biological sequences in the database and a comparison routine compares and scores a subject sequence against each known sequence in the comparison database as discussed by the authors.
Posted Content
Parrotron: An End-to-End Speech-to-Speech Conversion Model and its Applications to Hearing-Impaired Speech and Speech Separation
TL;DR: It is demonstrated that this model can be trained to normalize speech from any speaker regardless of accent, prosody, and background noise, into the voice of a single canonical target speaker with a fixed accent and consistent articulation and prosody.
Proceedings Article
Discriminative Topic Segmentation of Text and Speech
TL;DR: Two new discriminative topic segmentation algorithms are given which employ a new measure of text similarity based on word co-occurrence and it is demonstrated that by using a lattice of competing hypotheses rather than just the one-best hypothesis as input to the segmentation algorithm, the performance of the algorithm can be improved.
Journal ArticleDOI
Google USM: Scaling Automatic Speech Recognition Beyond 100 Languages
Yu Zhang,Wei Han,James Qin,Yong Wang,Ankur Bapna,Zhehuai Chen,Nanxin Chen,Bo Li,Vera Axelrod,Gary Wang,Zhongyao Meng,Ke Hu,Andrew Rosenberg,Rohit Prabhavalkar,Daniel S. Park,Parisa Haghani,Jason Riesa,Ginger Perng,Hagen Soltau,Trevor Strohman,Bhuvana Ramabhadran,Tara N. Sainath,Pedro J. Moreno,Chung-Cheng Chiu,Johan Schalkwyk,Francoise Beaufays,Yonghui Wu +26 more
TL;DR: Universal Speech Model (USM) as mentioned in this paper pre-trains the encoder of the model on a large unlabeled multilingual dataset of 12 million (M) hours spanning over 300 languages, and fine-tunes on a smaller labeled dataset.
Proceedings ArticleDOI
Selection and combination of hypotheses for dialectal speech recognition
TL;DR: This paper presents two methods to select and combine the best decoded hypothesis from a pool of dialectal recognizers, following a Machine Learning approach and extracts features from the Speech Recognition output along with Word Embeddings and use Shallow Neural Networks for classification.