Ian Lane

Yifan Peng, Siddharth Dalmia, Ian Lane et al. · cmu, nvidia

Conformer has proven to be effective in many speech processing tasks. It combines the benefits of extracting local dependencies using convolutions and global dependencies using self-attention. Inspired by this, we propose a more flexible, interpretable and customizable encoder alternative, Branchformer, with parallel branches for modeling various ranged dependencies in end-to-end speech processing. In each encoder layer, one branch employs self-attention or its variant to capture long-range dependencies, while the other branch utilizes an MLP module with convolutional gating (cgMLP) to extract local relationships. We conduct experiments on several speech recognition and spoken language understanding benchmarks. Results show that our model outperforms both Transformer and cgMLP. It also matches with or outperforms state-of-the-art results achieved by Conformer. Furthermore, we show various strategies to reduce computation thanks to the two-branch architecture, including the ability to have variable inference complexity in a single trained model. The weights learned for merging branches indicate how local and global dependencies are utilized in different layers, which benefits model designing.

Muqiao Yang, Ian Lane, Shinji Watanabe

Continual Learning, also known as Lifelong Learning, aims to continually learn from new data as it becomes available. While prior research on continual learning in automatic speech recognition has focused on the adaptation of models across multiple different speech recognition tasks, in this paper we propose an experimental setting for \textit{online continual learning} for automatic speech recognition of a single task. Specifically focusing on the case where additional training data for the same task becomes available incrementally over time, we demonstrate the effectiveness of performing incremental model updates to end-to-end speech recognition models with an online Gradient Episodic Memory (GEM) method. Moreover, we show that with online continual learning and a selective sampling strategy, we can maintain an accuracy that is similar to retraining a model from scratch while requiring significantly lower computation costs. We have also verified our method with self-supervised learning (SSL) features.

Priyesh Vakharia, Abigail Kufeldt, Max Meyers et al. · mit

Neurosymbolic approaches can add robustness to opaque neural systems by incorporating explainable symbolic representations. However, previous approaches have not used formal logic to contextualize queries to and validate outputs of large language models (LLMs). We propose \systemname{}, a novel neurosymbolic framework, to improve the robustness and reliability of LLMs in question-answering tasks. We provide \systemname{} with a domain-specific knowledge base, a logical reasoning system, and an integration to an existing LLM. This framework has two capabilities (1) context gathering: generating explainable and relevant context for a given query, and (2) validation: confirming and validating the factual accuracy of a statement in accordance with a knowledge base (KB). Our work opens a new area of neurosymbolic generative AI text validation and user personalization.

Sicong Huang, Jincheng He, Shiyuan Huang et al.

Hallucinations pose a significant challenge for large language models when answering knowledge-intensive queries. As LLMs become more widely adopted, it is crucial not only to detect if hallucinations occur but also to pinpoint exactly where in the LLM output they occur. SemEval 2025 Task 3, Mu-SHROOM: Multilingual Shared-task on Hallucinations and Related Observable Overgeneration Mistakes, is a recent effort in this direction. This paper describes the UCSC system submission to the shared Mu-SHROOM task. We introduce a framework that first retrieves relevant context, next identifies false content from the answer, and finally maps them back to spans in the LLM output. The process is further enhanced by automatically optimizing prompts. Our system achieves the highest overall performance, ranking #1 in average position across all languages. We release our code and experiment results.

Sicong Huang, Qianqi Yan, Shengze Wang et al.

Abstractive summarization using large language models (LLMs) has become an essential tool for condensing information. However, despite their ability to generate fluent summaries, these models sometimes produce unfaithful summaries, introducing hallucinations at the word, phrase, or concept level. Existing mitigation strategies, such as post-processing corrections or contrastive learning with synthetically generated negative samples, fail to fully address the diverse errors that can occur in LLM-generated summaries. In this paper, we investigate fine-tuning strategies to reduce the occurrence of unfaithful spans in generated summaries. First, we automatically generate summaries for the set of source documents in the training set with a variety of LLMs and then use GPT-4o to annotate any hallucinations it detects at the span-level. Leveraging these annotations, we fine-tune LLMs with both hallucination-free summaries and annotated unfaithful spans to enhance model faithfulness. In this paper, we introduce a new dataset that contains both faithful and unfaithful summaries with span-level labels and we evaluate three techniques to fine-tuning a LLM to improve the faithfulness of the resulting summarization: gradient ascent, unlikelihood training, and task vector negation. Experimental results show that all three approaches successfully leverage span-level annotations to improve faithfulness, with unlikelihood training being the most effective.

Benjamin Elizalde, Radu Revutchi, Samarjit Das et al.

In Psychology, actions are paramount for humans to identify sound events. In Machine Learning (ML), action recognition achieves high accuracy; however, it has not been asked whether identifying actions can benefit Sound Event Classification (SEC), as opposed to mapping the audio directly to a sound event. Therefore, we propose a new Psychology-inspired approach for SEC that includes identification of actions via human listeners. To achieve this goal, we used crowdsourcing to have listeners identify 20 actions that in isolation or in combination may have produced any of the 50 sound events in the well-studied dataset ESC-50. The resulting annotations for each audio recording relate actions to a database of sound events for the first time. The annotations were used to create semantic representations called Action Vectors (AVs). We evaluated SEC by comparing the AVs with two types of audio features -- log-mel spectrograms and state-of-the-art audio embeddings. Because audio features and AVs capture different abstractions of the acoustic content, we combined them and achieved one of the highest reported accuracies (88%).

Guan-Lin Chao, Abhinav Rastogi, Semih Yavuz et al.

Understanding and conversing about dynamic scenes is one of the key capabilities of AI agents that navigate the environment and convey useful information to humans. Video question answering is a specific scenario of such AI-human interaction where an agent generates a natural language response to a question regarding the video of a dynamic scene. Incorporating features from multiple modalities, which often provide supplementary information, is one of the challenging aspects of video question answering. Furthermore, a question often concerns only a small segment of the video, hence encoding the entire video sequence using a recurrent neural network is not computationally efficient. Our proposed question-guided video representation module efficiently generates the token-level video summary guided by each word in the question. The learned representations are then fused with the question to generate the answer. Through empirical evaluation on the Audio Visual Scene-aware Dialog (AVSD) dataset, our proposed models in single-turn and multi-turn question answering achieve state-of-the-art performance on several automatic natural language generation evaluation metrics.

Guan-Lin Chao, Ian Lane

An important yet rarely tackled problem in dialogue state tracking (DST) is scalability for dynamic ontology (e.g., movie, restaurant) and unseen slot values. We focus on a specific condition, where the ontology is unknown to the state tracker, but the target slot value (except for none and dontcare), possibly unseen during training, can be found as word segment in the dialogue context. Prior approaches often rely on candidate generation from n-gram enumeration or slot tagger outputs, which can be inefficient or suffer from error propagation. We propose BERT-DST, an end-to-end dialogue state tracker which directly extracts slot values from the dialogue context. We use BERT as dialogue context encoder whose contextualized language representations are suitable for scalable DST to identify slot values from their semantic context. Furthermore, we employ encoder parameter sharing across all slots with two advantages: (1) Number of parameters does not grow linearly with the ontology. (2) Language representation knowledge can be transferred among slots. Empirical evaluation shows BERT-DST with cross-slot parameter sharing outperforms prior work on the benchmark scalable DST datasets Sim-M and Sim-R, and achieves competitive performance on the standard DSTC2 and WOZ 2.0 datasets.

Guan-Lin Chao, William Chan, Ian Lane

Speech recognition in cocktail-party environments remains a significant challenge for state-of-the-art speech recognition systems, as it is extremely difficult to extract an acoustic signal of an individual speaker from a background of overlapping speech with similar frequency and temporal characteristics. We propose the use of speaker-targeted acoustic and audio-visual models for this task. We complement the acoustic features in a hybrid DNN-HMM model with information of the target speaker's identity as well as visual features from the mouth region of the target speaker. Experimentation was performed using simulated cocktail-party data generated from the GRID audio-visual corpus by overlapping two speakers's speech on a single acoustic channel. Our audio-only baseline achieved a WER of 26.3%. The audio-visual model improved the WER to 4.4%. Introducing speaker identity information had an even more pronounced effect, improving the WER to 3.6%. Combining both approaches, however, did not significantly improve performance further. Our work demonstrates that speaker-targeted models can significantly improve the speech recognition in cocktail party environments.

Tae Jin Park, Kyu Han, Ian Lane et al.

This work presents a novel approach to leverage lexical information for speaker diarization. We introduce a speaker diarization system that can directly integrate lexical as well as acoustic information into a speaker clustering process. Thus, we propose an adjacency matrix integration technique to integrate word level speaker turn probabilities with speaker embeddings in a comprehensive way. Our proposed method works without any reference transcript. Words, and word boundary information are provided by an ASR system. We show that our proposed method improves a baseline speaker diarization system solely based on speaker embeddings, achieving a meaningful improvement on the CALLHOME American English Speech dataset.

Ming Zeng, Haoxiang Gao, Tong Yu et al.

Deep neural networks, including recurrent networks, have been successfully applied to human activity recognition. Unfortunately, the final representation learned by recurrent networks might encode some noise (irrelevant signal components, unimportant sensor modalities, etc.). Besides, it is difficult to interpret the recurrent networks to gain insight into the models' behavior. To address these issues, we propose two attention models for human activity recognition: temporal attention and sensor attention. These two mechanisms adaptively focus on important signals and sensor modalities. To further improve the understandability and mean F1 score, we add continuity constraints, considering that continuous sensor signals are more robust than discrete ones. We evaluate the approaches on three datasets and obtain state-of-the-art results. Furthermore, qualitative analysis shows that the attention learned by the models agree well with human intuition.

Bing Liu, Ian Lane

In this work, we propose an adversarial learning method for reward estimation in reinforcement learning (RL) based task-oriented dialog models. Most of the current RL based task-oriented dialog systems require the access to a reward signal from either user feedback or user ratings. Such user ratings, however, may not always be consistent or available in practice. Furthermore, online dialog policy learning with RL typically requires a large number of queries to users, suffering from sample efficiency problem. To address these challenges, we propose an adversarial learning method to learn dialog rewards directly from dialog samples. Such rewards are further used to optimize the dialog policy with policy gradient based RL. In the evaluation in a restaurant search domain, we show that the proposed adversarial dialog learning method achieves advanced dialog success rate comparing to strong baseline methods. We further discuss the covariate shift problem in online adversarial dialog learning and show how we can address that with partial access to user feedback.

Ming Zeng, Tong Yu, Xiao Wang et al.

Labeled data used for training activity recognition classifiers are usually limited in terms of size and diversity. Thus, the learned model may not generalize well when used in real-world use cases. Semi-supervised learning augments labeled examples with unlabeled examples, often resulting in improved performance. However, the semi-supervised methods studied in the activity recognition literatures assume that feature engineering is already done. In this paper, we lift this assumption and present two semi-supervised methods based on convolutional neural networks (CNNs) to learn discriminative hidden features. Our semi-supervised CNNs learn from both labeled and unlabeled data while also performing feature learning on raw sensor data. In experiments on three real world datasets, we show that our CNNs outperform supervised methods and traditional semi-supervised learning methods by up to 18% in mean F1-score (Fm).

Kyu J. Han, Akshay Chandrashekaran, Jungsuk Kim et al.

In this paper we show how we have achieved the state-of-the-art performance on the industry-standard NIST 2000 Hub5 English evaluation set. We explore densely connected LSTMs, inspired by the densely connected convolutional networks recently introduced for image classification tasks. We also propose an acoustic model adaptation scheme that simply averages the parameters of a seed neural network acoustic model and its adapted version. This method was applied with the CallHome training corpus and improved individual system performances by on average 6.1% (relative) against the CallHome portion of the evaluation set with no performance loss on the Switchboard portion. With RNN-LM rescoring and lattice combination on the 5 systems trained across three different phone sets, our 2017 speech recognition system has obtained 5.0% and 9.1% on Switchboard and CallHome, respectively, both of which are the best word error rates reported thus far. According to IBM in their latest work to compare human and machine transcriptions, our reported Switchboard word error rate can be considered to surpass the human parity (5.1%) of transcribing conversational telephone speech.

Bing Liu, Ian Lane

The goal of this paper is to learn cross-domain representations for slot filling task in spoken language understanding (SLU). Most of the recently published SLU models are domain-specific ones that work on individual task domains. Annotating data for each individual task domain is both financially costly and non-scalable. In this work, we propose an adversarial training method in learning common features and representations that can be shared across multiple domains. Model that produces such shared representations can be combined with models trained on individual domain SLU data to reduce the amount of training samples required for developing a new domain. In our experiments using data sets from multiple domains, we show that adversarial training helps in learning better domain-general SLU models, leading to improved slot filling F1 scores. We further show that applying adversarial learning on domain-general model also helps in achieving higher slot filling performance when the model is jointly optimized with domain-specific models.

Bing Liu, Tong Yu, Ian Lane et al.

Dialog response selection is an important step towards natural response generation in conversational agents. Existing work on neural conversational models mainly focuses on offline supervised learning using a large set of context-response pairs. In this paper, we focus on online learning of response selection in retrieval-based dialog systems. We propose a contextual multi-armed bandit model with a nonlinear reward function that uses distributed representation of text for online response selection. A bidirectional LSTM is used to produce the distributed representations of dialog context and responses, which serve as the input to a contextual bandit. In learning the bandit, we propose a customized Thompson sampling method that is applied to a polynomial feature space in approximating the reward. Experimental results on the Ubuntu Dialogue Corpus demonstrate significant performance gains of the proposed method over conventional linear contextual bandits. Moreover, we report encouraging response selection performance of the proposed neural bandit model using the Recall@k metric for a small set of online training samples.

Bing Liu, Ian Lane

In this paper, we present a deep reinforcement learning (RL) framework for iterative dialog policy optimization in end-to-end task-oriented dialog systems. Popular approaches in learning dialog policy with RL include letting a dialog agent to learn against a user simulator. Building a reliable user simulator, however, is not trivial, often as difficult as building a good dialog agent. We address this challenge by jointly optimizing the dialog agent and the user simulator with deep RL by simulating dialogs between the two agents. We first bootstrap a basic dialog agent and a basic user simulator by learning directly from dialog corpora with supervised training. We then improve them further by letting the two agents to conduct task-oriented dialogs and iteratively optimizing their policies with deep RL. Both the dialog agent and the user simulator are designed with neural network models that can be trained end-to-end. Our experiment results show that the proposed method leads to promising improvements on task success rate and total task reward comparing to supervised training and single-agent RL training baseline models.

Bing Liu, Ian Lane

We present a novel end-to-end trainable neural network model for task-oriented dialog systems. The model is able to track dialog state, issue API calls to knowledge base (KB), and incorporate structured KB query results into system responses to successfully complete task-oriented dialogs. The proposed model produces well-structured system responses by jointly learning belief tracking and KB result processing conditioning on the dialog history. We evaluate the model in a restaurant search domain using a dataset that is converted from the second Dialog State Tracking Challenge (DSTC2) corpus. Experiment results show that the proposed model can robustly track dialog state given the dialog history. Moreover, our model demonstrates promising results in producing appropriate system responses, outperforming prior end-to-end trainable neural network models using per-response accuracy evaluation metrics.

Bing Liu, Ian Lane

In this work, we propose contextual language models that incorporate dialog level discourse information into language modeling. Previous works on contextual language model treat preceding utterances as a sequence of inputs, without considering dialog interactions. We design recurrent neural network (RNN) based contextual language models that specially track the interactions between speakers in a dialog. Experiment results on Switchboard Dialog Act Corpus show that the proposed model outperforms conventional single turn based RNN language model by 3.3% on perplexity. The proposed models also demonstrate advantageous performance over other competitive contextual language models.

Benjamin Elizalde, Ankit Shah, Siddharth Dalmia et al.

Audio Event Detection (AED) aims to recognize sounds within audio and video recordings. AED employs machine learning algorithms commonly trained and tested on annotated datasets. However, available datasets are limited in number of samples and hence it is difficult to model acoustic diversity. Therefore, we propose combining labeled audio from a dataset and unlabeled audio from the web to improve the sound models. The audio event detectors are trained on the labeled audio and ran on the unlabeled audio downloaded from YouTube. Whenever the detectors recognized any of the known sounds with high confidence, the unlabeled audio was use to re-train the detectors. The performance of the re-trained detectors is compared to the one from the original detectors using the annotated test set. Results showed an improvement of the AED, and uncovered challenges of using web audio from videos.

Bing Liu, Ian Lane

Speaker intent detection and semantic slot filling are two critical tasks in spoken language understanding (SLU) for dialogue systems. In this paper, we describe a recurrent neural network (RNN) model that jointly performs intent detection, slot filling, and language modeling. The neural network model keeps updating the intent estimation as word in the transcribed utterance arrives and uses it as contextual features in the joint model. Evaluation of the language model and online SLU model is made on the ATIS benchmarking data set. On language modeling task, our joint model achieves 11.8% relative reduction on perplexity comparing to the independent training language model. On SLU tasks, our joint model outperforms the independent task training model by 22.3% on intent detection error rate, with slight degradation on slot filling F1 score. The joint model also shows advantageous performance in the realistic ASR settings with noisy speech input.

Bing Liu, Ian Lane

Attention-based encoder-decoder neural network models have recently shown promising results in machine translation and speech recognition. In this work, we propose an attention-based neural network model for joint intent detection and slot filling, both of which are critical steps for many speech understanding and dialog systems. Unlike in machine translation and speech recognition, alignment is explicit in slot filling. We explore different strategies in incorporating this alignment information to the encoder-decoder framework. Learning from the attention mechanism in encoder-decoder model, we further propose introducing attention to the alignment-based RNN models. Such attentions provide additional information to the intent classification and slot label prediction. Our independent task models achieve state-of-the-art intent detection error rate and slot filling F1 score on the benchmark ATIS task. Our joint training model further obtains 0.56% absolute (23.8% relative) error reduction on intent detection and 0.23% absolute gain on slot filling over the independent task models.

Benjamin Elizalde, Anurag Kumar, Ankit Shah et al.

In this paper we present our work on Task 1 Acoustic Scene Classi- fication and Task 3 Sound Event Detection in Real Life Recordings. Among our experiments we have low-level and high-level features, classifier optimization and other heuristics specific to each task. Our performance for both tasks improved the baseline from DCASE: for Task 1 we achieved an overall accuracy of 78.9% compared to the baseline of 72.6% and for Task 3 we achieved a Segment-Based Error Rate of 0.76 compared to the baseline of 0.91.

Sebastian Sager, Benjamin Elizalde, Damian Borth et al.

Recently, sound recognition has been used to identify sounds, such as car and river. However, sounds have nuances that may be better described by adjective-noun pairs such as slow car, and verb-noun pairs such as flying insects, which are under explored. Therefore, in this work we investigate the relation between audio content and both adjective-noun pairs and verb-noun pairs. Due to the lack of datasets with these kinds of annotations, we collected and processed the AudioPairBank corpus consisting of a combined total of 1,123 pairs and over 33,000 audio files. One contribution is the previously unavailable documentation of the challenges and implications of collecting audio recordings with these type of labels. A second contribution is to show the degree of correlation between the audio content and the labels through sound recognition experiments, which yielded results of 70% accuracy, hence also providing a performance benchmark. The results and study in this paper encourage further exploration of the nuances in audio and are meant to complement similar research performed on images and text in multimedia analysis.

Benjamin Elizalde, Guan-Lin Chao, Ming Zeng et al.

City-identification of videos aims to determine the likelihood of a video belonging to a set of cities. In this paper, we present an approach using only audio, thus we do not use any additional modality such as images, user-tags or geo-tags. In this manner, we show to what extent the city-location of videos correlates to their acoustic information. Success in this task suggests improvements can be made to complement the other modalities. In particular, we present a method to compute and use semantic acoustic features to perform city-identification and the features show semantic evidence of the identification. The semantic evidence is given by a taxonomy of urban sounds and expresses the potential presence of these sounds in the city- soundtracks. We used the MediaEval Placing Task set, which contains Flickr videos labeled by city. In addition, we used the UrbanSound8K set containing audio clips labeled by sound- type. Our method improved the state-of-the-art performance and provides a novel semantic approach to this task

Suyoun Kim, Bhiksha Raj, Ian Lane

We propose a novel deep neural network architecture for speech recognition that explicitly employs knowledge of the background environmental noise within a deep neural network acoustic model. A deep neural network is used to predict the acoustic environment in which the system in being used. The discriminative embedding generated at the bottleneck layer of this network is then concatenated with traditional acoustic features as input to a deep neural network acoustic model. Through a series of experiments on Resource Management, CHiME-3 task, and Aurora4, we show that the proposed approach significantly improves speech recognition accuracy in noisy and highly reverberant environments, outperforming multi-condition training, noise-aware training, i-vector framework, and multi-task learning on both in-domain noise and unseen noise.

Suyoun Kim, Ian Lane

Integration of multiple microphone data is one of the key ways to achieve robust speech recognition in noisy environments or when the speaker is located at some distance from the input device. Signal processing techniques such as beamforming are widely used to extract a speech signal of interest from background noise. These techniques, however, are highly dependent on prior spatial information about the microphones and the environment in which the system is being used. In this work, we present a neural attention network that directly combines multi-channel audio to generate phonetic states without requiring any prior knowledge of the microphone layout or any explicit signal preprocessing for speech enhancement. We embed an attention mechanism within a Recurrent Neural Network (RNN) based acoustic model to automatically tune its attention to a more reliable input source. Unlike traditional multi-channel preprocessing, our system can be optimized towards the desired output in one step. Although attention-based models have recently achieved impressive results on sequence-to-sequence learning, no attention mechanisms have previously been applied to learn potentially asynchronous and non-stationary multiple inputs. We evaluate our neural attention model on the CHiME-3 challenge task, and show that the model achieves comparable performance to beamforming using a purely data-driven method.

William Chan, Nan Rosemary Ke, Ian Lane

Deep Neural Network (DNN) acoustic models have yielded many state-of-the-art results in Automatic Speech Recognition (ASR) tasks. More recently, Recurrent Neural Network (RNN) models have been shown to outperform DNNs counterparts. However, state-of-the-art DNN and RNN models tend to be impractical to deploy on embedded systems with limited computational capacity. Traditionally, the approach for embedded platforms is to either train a small DNN directly, or to train a small DNN that learns the output distribution of a large DNN. In this paper, we utilize a state-of-the-art RNN to transfer knowledge to small DNN. We use the RNN model to generate soft alignments and minimize the Kullback-Leibler divergence against the small DNN. The small DNN trained on the soft RNN alignments achieved a 3.93 WER on the Wall Street Journal (WSJ) eval92 task compared to a baseline 4.54 WER or more than 13% relative improvement.

William Chan, Ian Lane

We present a novel deep Recurrent Neural Network (RNN) model for acoustic modelling in Automatic Speech Recognition (ASR). We term our contribution as a TC-DNN-BLSTM-DNN model, the model combines a Deep Neural Network (DNN) with Time Convolution (TC), followed by a Bidirectional Long Short-Term Memory (BLSTM), and a final DNN. The first DNN acts as a feature processor to our model, the BLSTM then generates a context from the sequence acoustic signal, and the final DNN takes the context and models the posterior probabilities of the acoustic states. We achieve a 3.47 WER on the Wall Street Journal (WSJ) eval92 task or more than 8% relative improvement over the baseline DNN models.