Xuedong Zhang

Laxmi Pandey, Debjyoti Paul, Pooja Chitkara et al. · meta-ai

Inverse text normalization (ITN) is used to convert the spoken form output of an automatic speech recognition (ASR) system to a written form. Traditional handcrafted ITN rules can be complex to transcribe and maintain. Meanwhile neural modeling approaches require quality large-scale spoken-written pair examples in the same or similar domain as the ASR system (in-domain data), to train. Both these approaches require costly and complex annotations. In this paper, we present a data augmentation technique that effectively generates rich spoken-written numeric pairs from out-of-domain textual data with minimal human annotation. We empirically demonstrate that ITN model trained using our data augmentation technique consistently outperform ITN model trained using only in-domain data across all numeric surfaces like cardinal, currency, and fraction, by an overall accuracy of 14.44%.

Sen Fang, Chunyu Sui, Yanghao Zhou et al.

In this paper, we propose a dual-condition diffusion pre-training model named SignDiff that can generate human sign language speakers from a skeleton pose. SignDiff has a novel Frame Reinforcement Network called FR-Net, similar to dense human pose estimation work, which enhances the correspondence between text lexical symbols and sign language dense pose frames, reduces the occurrence of multiple fingers in the diffusion model. In addition, we propose a new method for American Sign Language Production (ASLP), which can generate ASL skeletal pose videos from text input, integrating two new improved modules and a new loss function to improve the accuracy and quality of sign language skeletal posture and enhance the ability of the model to train on large-scale data. We propose the first baseline for ASL production and report the scores of 17.19 and 12.85 on BLEU-4 on the How2Sign dev/test sets. We evaluated our model on the previous mainstream dataset PHOENIX14T, and the experiments achieved the SOTA results. In addition, our image quality far exceeds all previous results by 10 percentage points in terms of SSIM.

Zhe Liu, Xuedong Zhang, Fuchun Peng

Recent research has shown that language models have a tendency to memorize rare or unique sequences in the training corpora which can thus leak sensitive attributes of user data. We employ a teacher-student framework and propose a novel approach called alternating teaching to mitigate unintended memorization in sequential modeling. In our method, multiple teachers are trained on disjoint training sets whose privacy one wishes to protect, and teachers' predictions supervise the training of a student model in an alternating manner at each time step. Experiments on LibriSpeech datasets show that the proposed method achieves superior privacy-preserving results than other counterparts. In comparison with no prevention for unintended memorization, the overall utility loss is small when training records are sufficient.

Laxmi Pandey, Ke Li, Jinxi Guo et al.

Multilingual pretraining for transfer learning significantly boosts the robustness of low-resource monolingual ASR models. This study systematically investigates three main aspects: (a) the impact of transfer learning on model performance during initial training or fine-tuning, (b) the influence of transfer learning across dataset domains and languages, and (c) the effect on rare-word recognition compared to non-rare words. Our finding suggests that RNNT-loss pretraining, followed by monolingual fine-tuning with Minimum Word Error Rate (MinWER) loss, consistently reduces Word Error Rates (WER) across languages like Italian and French. WER Reductions (WERR) reach 36.2% and 42.8% compared to monolingual baselines for MLS and in-house datasets. Out-of-domain pretraining leads to 28% higher WERR than in-domain pretraining. Both rare and non-rare words benefit, with rare words showing greater improvements with out-of-domain pretraining, and non-rare words with in-domain pretraining.

Yancheng Wang, Osama Hanna, Ruiming Xie et al.

Emotion recognition in speech presents a complex multimodal challenge, requiring comprehension of both linguistic content and vocal expressivity, particularly prosodic features such as fundamental frequency, intensity, and temporal dynamics. Although large language models (LLMs) have shown promise in reasoning over textual transcriptions for emotion recognition, they typically neglect fine-grained prosodic information, limiting their effectiveness and interpretability. In this work, we propose VowelPrompt, a linguistically grounded framework that augments LLM-based emotion recognition with interpretable, fine-grained vowel-level prosodic cues. Drawing on phonetic evidence that vowels serve as primary carriers of affective prosody, VowelPrompt extracts pitch-, energy-, and duration-based descriptors from time-aligned vowel segments, and converts these features into natural language descriptions for better interpretability. Such a design enables LLMs to jointly reason over lexical semantics and fine-grained prosodic variation. Moreover, we adopt a two-stage adaptation procedure comprising supervised fine-tuning (SFT) followed by Reinforcement Learning with Verifiable Reward (RLVR), implemented via Group Relative Policy Optimization (GRPO), to enhance reasoning capability, enforce structured output adherence, and improve generalization across domains and speaker variations. Extensive evaluations across diverse benchmark datasets demonstrate that VowelPrompt consistently outperforms state-of-the-art emotion recognition methods under zero-shot, fine-tuned, cross-domain, and cross-linguistic conditions, while enabling the generation of interpretable explanations that are jointly grounded in contextual semantics and fine-grained prosodic structure.

Yingyi Ma, Zhe Liu, Xuedong Zhang

Neural network language model (NNLM) plays an essential role in automatic speech recognition (ASR) systems, especially in adaptation tasks when text-only data is available. In practice, an NNLM is typically trained on a combination of data sampled from multiple corpora. Thus, the data sampling strategy is important to the adaptation performance. Most existing works focus on designing static sampling strategies. However, each corpus may show varying impacts at different NNLM training stages. In this paper, we introduce a novel adaptive multi-corpora training algorithm that dynamically learns and adjusts the sampling probability of each corpus along the training process. The algorithm is robust to corpora sizes and domain relevance. Compared with static sampling strategy baselines, the proposed approach yields remarkable improvement by achieving up to relative 7% and 9% word error rate (WER) reductions on in-domain and out-of-domain adaptation tasks, respectively.

Szu-Jui Chen, Debjyoti Paul, Yutong Pang et al.

With the emergence of automatic speech recognition (ASR) models, converting the spoken form text (from ASR) to the written form is in urgent need. This inverse text normalization (ITN) problem attracts the attention of researchers from various fields. Recently, several works show that data-driven ITN methods can output high-quality written form text. Due to the scarcity of labeled spoken-written datasets, the studies on non-English data-driven ITN are quite limited. In this work, we propose a language-agnostic data-driven ITN framework to fill this gap. Specifically, we leverage the data augmentation in conjunction with neural machine translated data for low resource languages. Moreover, we design an evaluation method for language agnostic ITN model when only English data is available. Our empirical evaluation shows this language agnostic modeling approach is effective for low resource languages while preserving the performance for high resource languages.

Yutong Pang, Debjyoti Paul, Kevin Jiang et al.

This paper introduces two advancements in the field of Large Language Model Annotation with a focus on punctuation restoration tasks. Our first contribution is the application of LLaMA for punctuation restoration, which demonstrates superior performance compared to the established benchmark. Despite its impressive quality, LLaMA faces challenges regarding inference speed and hallucinations. To address this, our second contribution presents Forward Pass Only Decoding (FPOD), a novel decoding approach for annotation tasks. This innovative method results in a substantial 19.8x improvement in inference speed, effectively addressing a critical bottleneck and enhancing the practical utility of LLaMA for large-scale data annotation tasks without hallucinations. The combination of these contributions not only solidifies LLaMA as a powerful tool for punctuation restoration but also highlights FPOD as a crucial strategy for overcoming speed constraints.

Yong Ma, Xuesong Zhang, Xuedong Zhang et al.

Voice assistants (VAs) are typically evaluated through task performance metrics and self-report questionnaires, but people's voices themselves carry rich paralinguistic cues that reveal affect, effort, and interaction breakdowns. We present a within-subjects study (N=49) that systematically compared three VA personas across three usage scenarios to investigate whether speech-derived audio features can serve as a proxy for user experience (UX). Participants' speech was analyzed for temporal, spectral, and linguistic markers, alongside standardized UX measures, brief mood and stress ratings, and a post-study questionnaire. We found correlations between specific speech features and self-reported satisfaction and experience. Furthermore, a machine learning model trained on speech features achieved promising accuracy in classifying UX levels, indicating that this might be a reasonable alternative to self-report instruments. Our findings establish speech as a viable, real-time signal for implicitly measuring UX and point toward adaptive VUIs that respond dynamically to emotional and usability-related vocal cues.

Chung-Cheng Chiu, Anshuman Tripathi, Katherine Chou et al.

In this work we explored building automatic speech recognition models for transcribing doctor patient conversation. We collected a large scale dataset of clinical conversations ($14,000$ hr), designed the task to represent the real word scenario, and explored several alignment approaches to iteratively improve data quality. We explored both CTC and LAS systems for building speech recognition models. The LAS was more resilient to noisy data and CTC required more data clean up. A detailed analysis is provided for understanding the performance for clinical tasks. Our analysis showed the speech recognition models performed well on important medical utterances, while errors occurred in causal conversations. Overall we believe the resulting models can provide reasonable quality in practice.