Zobrazeno 1 - 10
of 126
pro vyhledávání: '"Salazar, Julian"'
Autor:
Battenberg, Eric, Skerry-Ryan, RJ, Stanton, Daisy, Mariooryad, Soroosh, Shannon, Matt, Salazar, Julian, Kao, David
Autoregressive (AR) Transformer-based sequence models are known to have difficulty generalizing to sequences longer than those seen during training. When applied to text-to-speech (TTS), these models tend to drop or repeat words or produce erratic ou
Externí odkaz:
http://arxiv.org/abs/2410.22179
Autor:
Nachmani, Eliya, Levkovitch, Alon, Hirsch, Roy, Salazar, Julian, Asawaroengchai, Chulayuth, Mariooryad, Soroosh, Rivlin, Ehud, Skerry-Ryan, RJ, Ramanovich, Michelle Tadmor
We present Spectron, a novel approach to adapting pre-trained large language models (LLMs) to perform spoken question answering (QA) and speech continuation. By endowing the LLM with a pre-trained speech encoder, our model becomes able to take speech
Externí odkaz:
http://arxiv.org/abs/2305.15255
End-to-end (E2E) spoken language understanding (SLU) is constrained by the cost of collecting speech-semantics pairs, especially when label domains change. Hence, we explore \textit{zero-shot} E2E SLU, which learns E2E SLU without speech-semantics pa
Externí odkaz:
http://arxiv.org/abs/2305.12793
Large pretrained language models (PLMs) are often domain- or task-adapted via fine-tuning or prompting. Finetuning requires modifying all of the parameters and having enough data to avoid overfitting while prompting requires no training and few examp
Externí odkaz:
http://arxiv.org/abs/2207.03509
Autor:
Salvi, Valeria Franco, Salazar, Julián, Lillo, Jordi A. López, Fiorani, Agustina Vázquez, Montegú, Juan
Publikováno v:
Estudios Atacameños, 2023 Jan 01. 69, 1-35.
Externí odkaz:
https://www.jstor.org/stable/48772066
Publikováno v:
In Journal of Archaeological Science: Reports May 2024 55
Non-autoregressive models greatly improve decoding speed over typical sequence-to-sequence models, but suffer from degraded performance. Infilling and iterative refinement models make up some of this gap by editing the outputs of a non-autoregressive
Externí odkaz:
http://arxiv.org/abs/2010.14233
We describe an unsupervised method to create pseudo-parallel corpora for machine translation (MT) from unaligned text. We use multilingual BERT to create source and target sentence embeddings for nearest-neighbor search and adapt the model via self-t
Externí odkaz:
http://arxiv.org/abs/2010.07761
Multilingual contextual embeddings have demonstrated state-of-the-art performance in zero-shot cross-lingual transfer learning, where multilingual BERT is fine-tuned on one source language and evaluated on a different target language. However, publis
Externí odkaz:
http://arxiv.org/abs/2004.15001
We discuss the problem of echographic transcription in autoregressive sequence-to-sequence attentional architectures for automatic speech recognition, where a model produces very long sequences of repetitive outputs when presented with out-of-domain
Externí odkaz:
http://arxiv.org/abs/2002.05150