English
Related papers

Related papers: A Paradigm for Interpreting Metrics and Identifyin…

200 papers

Training large foundation models using self-supervised objectives on unlabeled data, followed by fine-tuning on downstream tasks, has emerged as a standard procedure. Unfortunately, the efficacy of this approach is often constrained by both…

A measure of similarity between text embeddings can be considered adequate only if it adheres to the human perception of similarity between texts. In this paper, we introduce the distance-to-distance ratio (DDR), a novel measure of…

Computation and Language · Computer Science 2026-01-27 Abdullah Qureshi , Kenneth Rice , Alexander Wolpert

Despite speech recognition systems achieving low word error rates on standard benchmarks, they often fail on short, high-stakes utterances in real-world deployments. Here, we study this failure mode in a high-stakes task: the transcription…

Artificial Intelligence · Computer Science 2026-02-18 Kaitlyn Zhou , Martijn Bartelds , Federico Bianchi , James Zou

Word-level quality estimation (WQE) aims to automatically identify fine-grained error spans in machine-translated outputs and has found many uses, including assisting translators during post-editing. Modern WQE techniques are often…

Computation and Language · Computer Science 2025-11-18 Gabriele Sarti , Vilém Zouhar , Malvina Nissim , Arianna Bisazza

Automatic speech recognition (ASR) systems are primarily evaluated on transcription accuracy. However, in some use cases such as subtitling, verbatim transcription would reduce output readability given limited screen size and reading time.…

Computation and Language · Computer Science 2020-05-26 Danni Liu , Jan Niehues , Gerasimos Spanakis

In recent years, word embeddings have been widely used to measure biases in texts. Even if they have proven to be effective in detecting a wide variety of biases, metrics based on word embeddings lack transparency and interpretability. We…

Computation and Language · Computer Science 2023-07-19 Francisco Valentini , Germán Rosati , Damián Blasi , Diego Fernandez Slezak , Edgar Altszyler

Modern Automatic Speech Recognition (ASR) systems can achieve high performance in terms of recognition accuracy. However, a perfectly accurate transcript still can be challenging to read due to disfluency, filter words, and other errata…

Computation and Language · Computer Science 2021-02-23 Junwei Liao , Yu Shi , Ming Gong , Linjun Shou , Sefik Eskimez , Liyang Lu , Hong Qu , Michael Zeng

Automatic Speech Recognition (ASR) systems are commonly evaluated using aggregate metrics such as Word Error Rate (WER), which do not capture the linguistic structure of errors. Fine-grained analysis, such as Part-of-Speech (PoS)-wise error…

Computation and Language · Computer Science 2026-05-28 Prasenjit K Mudi , Dahlia Devapriya , Sheetal Kalyani

Automatic speech recognition systems have accomplished remarkable improvements in transcription accuracy in recent years. On some domains, models now achieve near-human performance. However, transcription performance on oral history has not…

Audio and Speech Processing · Electrical Eng. & Systems 2022-01-19 Michael Gref , Nike Matthiesen , Christoph Schmidt , Sven Behnke , Joachim Köhler

Automatic metrics are fundamental for the development and evaluation of machine translation systems. Judging whether, and to what extent, automatic metrics concur with the gold standard of human evaluation is not a straightforward problem.…

Computation and Language · Computer Science 2020-06-15 Nitika Mathur , Timothy Baldwin , Trevor Cohn

We introduce the SEER (Span-based Emotion Evidence Retrieval) Benchmark to test Large Language Models' (LLMs) ability to identify the specific spans of text that express emotion. Unlike traditional emotion recognition tasks that assign a…

Computation and Language · Computer Science 2025-10-29 Aneesha Sampath , Oya Aran , Emily Mower Provost

The issue of fairness arises when the automatic speech recognition (ASR) systems do not perform equally well for all subgroups of the population. In any fairness measurement studies for ASR, the open questions of how to control the nuisance…

Machine Learning · Statistics 2021-09-21 Zhe Liu , Irina-Elena Veliche , Fuchun Peng

One of the most important problems in machine translation (MT) evaluation is to evaluate the similarity between translation hypotheses with different surface forms from the reference, especially at the segment level. We propose to use word…

Computation and Language · Computer Science 2017-04-04 Junki Matsuo , Mamoru Komachi , Katsuhito Sudoh

Confidence estimate is an often requested feature in applications such as medical transcription where errors can impact patient care and the confidence estimate could be used to alert medical professionals to verify potential errors in…

Computation and Language · Computer Science 2021-10-29 Mingqiu Wang , Hagen Soltau , Laurent El Shafey , Izhak Shafran

Parameter-efficient fine-tuning (PEFT) for personalizing automatic speech recognition (ASR) has recently shown promise for adapting general population models to atypical speech. However, these approaches assume a priori knowledge of the…

Machine Learning · Computer Science 2024-07-04 Max Müller-Eberstein , Dianna Yee , Karren Yang , Gautam Varma Mantena , Colin Lea

Word embedding spaces are powerful tools for capturing latent semantic relationships between terms in corpora, and have become widely popular for building state-of-the-art natural language processing algorithms. However, studies have shown…

Computation and Language · Computer Science 2019-06-21 Inom Mirzaev , Anthony Schulte , Michael Conover , Sam Shah

We present a cost-effective approach for developing Automatic Speech Recognition (ASR) models for low-resource languages like Ika. We fine-tune the pretrained wav2vec 2.0 Massively Multilingual Speech Models on a high-quality speech dataset…

Computation and Language · Computer Science 2024-10-03 Uchenna Nzenwata , Daniel Ogbuigwe

Most current state-of-the art systems for generating English text from Abstract Meaning Representation (AMR) have been evaluated only using automated metrics, such as BLEU, which are known to be problematic for natural language generation.…

Computation and Language · Computer Science 2020-12-02 Emma Manning , Shira Wein , Nathan Schneider

This paper presents a novel evaluation approach to text-based speaker diarization (SD), tackling the limitations of traditional metrics that do not account for any contextual information in text. Two new metrics are proposed, Text-based…

Computation and Language · Computer Science 2023-09-15 Chen Gong , Peilin Wu , Jinho D. Choi

The transcription quality of automatic speech recognition (ASR) systems degrades significantly when transcribing audios coming from unseen domains. We propose an unsupervised error correction method for unsupervised ASR domain adaption,…

Sound · Computer Science 2022-09-27 Long Mai , Julie Carson-Berndsen