WER we are and WER we think we are

no code implementations • Findings of the Association for Computational Linguistics 2020 • Piotr Szymański, Piotr Żelasko, Mikolaj Morzy, Adrian Szymczak, Marzena Żyła-Hoppe, Joanna Banaszczak, Lukasz Augustyniak, Jan Mizgajski, Yishay Carmiel

Natural language processing of conversational speech requires the availability of high-quality transcripts.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +1

Punctuation Prediction in Spontaneous Conversations: Can We Mitigate ASR Errors with Retrofitted Word Embeddings?

no code implementations • 13 Apr 2020 • Łukasz Augustyniak, Piotr Szymanski, Mikołaj Morzy, Piotr .Zelasko, Adrian Szymczak, Jan Mizgajski, Yishay Carmiel, Najim Dehak

Automatic Speech Recognition (ASR) systems introduce word errors, which often confuse punctuation prediction models, turning punctuation restoration into a challenging task.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +3

Hierarchical Transformers for Long Document Classification

3 code implementations • 23 Oct 2019 • Raghavendra Pappagari, Piotr Żelasko, Jesús Villalba, Yishay Carmiel, Najim Dehak

BERT, which stands for Bidirectional Encoder Representations from Transformers, is a recently introduced language representation model based upon the transfer learning paradigm.

Classification Document Classification +3

Avaya Conversational Intelligence: A Real-Time System for Spoken Language Understanding in Human-Human Call Center Conversations

no code implementations • 2 Sep 2019 • Jan Mizgajski, Adrian Szymczak, Robert Głowski, Piotr Szymański, Piotr Żelasko, Łukasz Augustyniak, Mikołaj Morzy, Yishay Carmiel, Jeff Hodson, Łukasz Wójciak, Daniel Smoczyk, Adam Wróbel, Bartosz Borowik, Adam Artajew, Marcin Baran, Cezary Kwiatkowski, Marzena Żyła-Hoppe

Avaya Conversational Intelligence(ACI) is an end-to-end, cloud-based solution for real-time Spoken Language Understanding for call centers.

Abstractive Text Summarization Intent Recognition +4

Towards Better Understanding of Spontaneous Conversations: Overcoming Automatic Speech Recognition Errors With Intent Recognition

no code implementations • 21 Aug 2019 • Piotr Żelasko, Jan Mizgajski, Mikołaj Morzy, Adrian Szymczak, Piotr Szymański, Łukasz Augustyniak, Yishay Carmiel

In this paper, we present a method for correcting automatic speech recognition (ASR) errors using a finite state transducer (FST) intent recognition framework.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +3

Punctuation Prediction Model for Conversational Speech

no code implementations • 2 Jul 2018 • Piotr Żelasko, Piotr Szymański, Jan Mizgajski, Adrian Szymczak, Yishay Carmiel, Najim Dehak

The models are trained on the Fisher corpus which includes punctuation annotation.