This paper introduces GigaSpeech, an evolving, multi-domain English speech recognition corpus with 10,000 hours of high quality labeled audio suitable for supervised training, and 40,000 hours of total audio suitable for semi-supervised and unsupervised training.Around 40,000 hours of transcribed audio is first collected from audiobooks, podcasts and YouTube, covering both read and spontaneous speaking styles, and a variety of topics, such as arts, science, sports, etc.A new forced alignment and segmentation pipeline is proposed to create sentence segments suitable for speech recognition training, and to filter out segments with low-quality transcription.For system training, GigaSpeech provides five subsets of different sizes, 10h, 250h, 1000h, 2500h, and 10000h.For our 10,000-hour XL training subset, we cap the word error rate at 4% during the filtering/validation stage, and for all our other smaller training subsets, we cap it at 0%.The DEV and TEST evaluation sets, on the other hand, are re-processed by professional human transcribers to ensure high transcription quality.Baseline systems are provided for popular speech recognition toolkits, namely Athena, ESPnet, Kaldi and Pika.

For the task of question answering (QA) over Freebase on the WEBQUESTIONS dataset (Berant et al., 2013), we found that 85% of all questions (in the training set) can be directly answered via a single binary relation.Thus we turned this task into slot-filling for tuples: predicting relations to get answers given a question's topic.We design efficient data structures to identify question topics organically from 46 million Freebase topic names, without employing any NLP processing tools.Then we present a lean QA system that runs in real time (in offline batch testing it answered two thousand questions in 51 seconds on a laptop).The system also achieved 7.8% better F 1 score (harmonic mean of average precision and recall) than the previous state of the art.

Answering natural language questions using the Freebase knowledge base has recently been explored as a platform for advancing the state of the art in open domain semantic parsing.Those efforts map questions to sophisticated meaning representations that are then attempted to be matched against viable answer candidates in the knowledge base.Here we show that relatively modest information extraction techniques, when paired with a webscale corpus, can outperform these sophisticated approaches by roughly 34% relative gain.

We introduce a novel discriminative model for phrase-based monolingual alignment using a semi-Markov CRF.Our model achieves stateof-the-art alignment accuracy on two phrasebased alignment datasets (RTE and paraphrase), while doing significantly better than other strong baselines in both non-identical alignment and phrase-only alignment.Additional experiments highlight the potential benefit of our alignment model to RTE, paraphrase identification and question answering, where even a naive application of our model's alignment score approaches the state of the art.