2 |
Between words and characters: A Brief History of Open-Vocabulary Modeling and Tokenization in NLP
|
|
|
|
In: https://hal.inria.fr/hal-03540069 ; 2022 (2022)
|
|
BASE
|
|
Show details
|
|
3 |
The (white) ears of Ofsted: a raciolinguistic perspective on the listening practices of the schools inspectorate
|
|
|
|
BASE
|
|
Show details
|
|
5 |
Distinct neural signatures of schizotypy and psychopathy during visual word-nonword recognition
|
|
|
|
BASE
|
|
Show details
|
|
6 |
Application of Article 5 of the ECHR to the detention of a person who has committed a criminal offense
|
|
|
|
In: Revista de Direito Internacional; v. 19, n. 1 (2022): International Law and climate litigation ; 2237-1036 ; 2236-997X (2022)
|
|
BASE
|
|
Show details
|
|
7 |
Development of a standard of care for patients with valosin-containing protein associated multisystem proteinopathy.
|
|
|
|
In: Orphanet journal of rare diseases, vol 17, iss 1 (2022)
|
|
BASE
|
|
Show details
|
|
8 |
Traduction automatique et doublage : impressions d'une expérience d'enseignement
|
|
|
|
In: https://hal.archives-ouvertes.fr/hal-03583626 ; 2022 (2022)
|
|
BASE
|
|
Show details
|
|
9 |
Makoraba, Mochorba & Maka Revisited: A Geo-Linguistic Perspective ...
|
|
|
|
BASE
|
|
Show details
|
|
11 |
Protocol for the development of the international population registry for aphasia after stroke (I-PRAISE)
|
|
|
|
In: Research outputs 2014 to 2021 (2022)
|
|
BASE
|
|
Show details
|
|
12 |
SHAS: Approaching optimal Segmentation for End-to-End Speech Translation ...
|
|
|
|
Abstract:
Speech translation models are unable to directly process long audios, like TED talks, which have to be split into shorter segments. Speech translation datasets provide manual segmentations of the audios, which are not available in real-world scenarios, and existing segmentation methods usually significantly reduce translation quality at inference time. To bridge the gap between the manual segmentation of training and the automatic one at inference, we propose Supervised Hybrid Audio Segmentation (SHAS), a method that can effectively learn the optimal segmentation from any manually segmented speech corpus. First, we train a classifier to identify the included frames in a segmentation, using speech representations from a pre-trained wav2vec 2.0. The optimal splitting points are then found by a probabilistic Divide-and-Conquer algorithm that progressively splits at the frame of lowest probability until all segments are below a pre-specified length. Experiments on MuST-C and mTEDx show that the translation of ... : Submitted to Interspeech 2022, 5 pages. Previous version (v1) has additionally a 2-page Appendix ...
|
|
Keyword:
Audio and Speech Processing eess.AS; Computation and Language cs.CL; FOS Computer and information sciences; FOS Electrical engineering, electronic engineering, information engineering; Sound cs.SD
|
|
URL: https://arxiv.org/abs/2202.04774 https://dx.doi.org/10.48550/arxiv.2202.04774
|
|
BASE
|
|
Hide details
|
|
13 |
A computational investigation of inventive spelling and the “Lesen durch Schreiben” method ...
|
|
|
|
BASE
|
|
Show details
|
|
18 |
Characteristics of Auditory Processing Disorders (de Wit et al., 2016) ...
|
|
|
|
BASE
|
|
Show details
|
|
19 |
Characteristics of Auditory Processing Disorders (de Wit et al., 2016) ...
|
|
|
|
BASE
|
|
Show details
|
|
20 |
Working memory predicts word learning (Gray et al., 2022) ...
|
|
|
|
BASE
|
|
Show details
|
|
|
|