arXiv subjects

Raul Monteiro

Publications and source records attributed to Raul Monteiro.

3 recordsLinked to original sources

arXiv2026

Massive Open-Vocabulary Keyword Spotting

Automatic speech recognition systems have been shown to under-perform when it comes to transcribing words rarely seen in the training data, namely specialized terminology. Open-vocabulary keyword spotting, combined with contextual biasing, has been shown to mitigate this issue. However, existing systems can only handle glossaries of a few hundred terms without becoming an infeasible bottleneck. We propose a system that stores features with a memory footprint up to 128 times smaller than a comparable baseline and allows users to process massive databases while remaining open-vocabulary. Without fine-tuning the speech recognition model, our system achieves a comparable entity recall as uncompressed solutions, even in languages not seen during training.

eess.AS↗

arXiv2023

Towards End-to-end Speech-to-text Summarization

Speech-to-text (S2T) summarization is a time-saving technique for filtering and keeping up with the broadcast news uploaded online on a daily basis. The rise of large language models from deep learning with impressive text generation capabilities has placed the research focus on summarization systems that produce paraphrased compact versions of the document content, also known as abstractive summaries. End-to-end (E2E) modelling of S2T abstractive summarization is a promising approach that offers the possibility of generating rich latent representations that leverage non-verbal and acoustic information, as opposed to the use of only linguistic information from automatically generated transcripts in cascade systems. However, the few literature on E2E modelling of this task fails on exploring different domains, namely broadcast news, which is challenging domain where large and diversified volumes of data are presented to the user every day. We model S2T summarization both with a cascade and an E2E system for a corpus of broadcast news in French. Our novel E2E model leverages external data by resorting to transfer learning from a pre-trained T2T summarizer. Experiments show that both our cascade and E2E abstractive summarizers are stronger than an extractive baseline. However, the performance of the E2E model still lies behind the cascade one, which is object of an extensive analysis that includes future directions to close that gap.

cs.CL↗

arXiv2023

Entanglement Routing Based on Fidelity Curves

How to efficiently distribute entanglement over large-scale quantum networks is still an open problem that greatly depends on the technology considered. In this work, we consider quantum networks where each link is characterized by a trade-off between the entanglement generation rate and fidelity. For such networks, we look at the two following problems: the one of finding the best path to connect any two given nodes, and the problem of finding the best starting node in order to connect three nodes in the network multipartite entanglement routing. Two entanglement distribution models are considered: one where entangled qubits are distributed one at a time, and a flow model where a large number of entangled qubits are distributed simultaneously. The paper proposes of a quite general methodology that uses continuous fidelity curves (i.e., entanglement generation fidelity vs. rate) as the main routing metric. Combined with multi-objective path-finding algorithms, the fidelity curves describing each link allow finding a set of paths that maximize both the end-to-end fidelity and the entanglement generation rate. For the link models and networks considered, it is proven that the algorithm always converges to the optimal solution. It is also shown through simulation that the execution time grows polynomially with the number of network nodes (growing with a power between $1$ and $1.4$, depending on the network)

quant-ph↗