• Corpus ID: 31813643

Meeting assistant application

  title={Meeting assistant application},
  author={Michel Assayag and Jonathan Huang and Jonathan Mamou and Oren Pereg and Saurav Sahay and Oren Shamir and Georg Stemmer and Moshe Wasserblat},
This paper describes the Meeting Assistant application developed at Intel. Unlike existing human-to-machine solutions, the challenges induced by human-to-human conversations are currently poorly addressed by the industry. In this paper, we describe the capabilities of in-house speech and NLP technologies: online automatic speech recognition, speaker diarization, keyphrase extraction and sentiment detection. These technologies has been adapted to conversational speech domain and integrated into… 

Figures from this paper



Analysis of i-vector Length Normalization in Speaker Recognition Systems

The proposed approach deals with the nonGaussian behavior of i-vectors by performing a simple length normalization, which allows the use of probabilistic models with Gaussian assumptions that yield equivalent performance to that of more complicated systems based on Heavy-Tailed assumptions.

The Kaldi Speech Recognition Toolkit

The design of Kaldi is described, a free, open-source toolkit for speech recognition research that provides a speech recognition system based on finite-state automata together with detailed documentation and a comprehensive set of scripts for building complete recognition systems.

Opinion Word Expansion and Target Extraction through Double Propagation

This article study two important problems, namely, opinion lexicon expansion and opinion target extraction, and proposes a method based on bootstrapping that outperforms these existing methods significantly.

Automatic Keyword Extraction from Individual Documents

Keywords, which we define as a sequence of one or more words, provide a compact representation of a document’s content. Ideally, keywords represent in condensed form the essential content of a

TermExtractor: a Web Application to Learn the Common Terminology of Interest Groups and Research Communities

A high-performing technique to automatically extract from the available documents in a given domain the shared terminology of a web community has been successfully experimented and submitted for large-scale evaluation in the domain of enterprise interoperability, by the member of the INTEROP network of excellence.

Front-End Factor Analysis for Speaker Verification

An extension of the previous work which proposes a new speaker representation for speaker verification, a new low-dimensional speaker- and channel-dependent space is defined using a simple factor analysis, named the total variability space because it models both speaker and channel variabilities.

The kaldi speech recognition toolkit , ” in IEEE 2011 Workshop on Automatic Speech Recognition and Understanding

  • IEEE Signal Processing Society , Dec .
  • 2011