We propose a method for learning de-identified prosody representations from raw audio using a contrastive self-supervised signal. Whereas prior work has relied on conditioning models on bottlenecks, we introduce a set of inductive biases that exploit the natural structure of prosody to minimize timbral information and decouple prosody from speaker representations. Despite aggressive downsampling of the input and having no access to linguistic information, our model performs comparably to state-of-the-art speech representations on DAMMP, a new benchmark we introduce for spoken language understanding. We use minimum description length probing to show that our representations have selectively learned the subcomponents of non-timbral prosody, and that the product quantizer naturally disentangles them without using bottlenecks. We derive an information-theoretic definition of speech de-identifiability and use it to demonstrate that our prosody representations are less identifiable than other speech representations.
Dialog acts can be interpreted as the atomic units of a conversation, more fine-grained than utterances, characterized by a specific communicative function. The ability to structure a conversational transcript as a sequence of dialog acts -- dialog act recognition, including the segmentation -- is critical for understanding dialog. We apply two pre-trained transformer models, XLNet and Longformer, to this task in English and achieve strong results on Switchboard Dialog Act and Meeting Recorder Dialog Act corpora with dialog act segmentation error rates (DSER) of 8.4% and 14.2%. To understand the key factors affecting dialog act recognition, we perform a comparative analysis of models trained under different conditions. We find that the inclusion of a broader conversational context helps disambiguate many dialog act classes, especially those infrequent in the training data. The presence of punctuation in the transcripts has a massive effect on the models' performance, and a detailed analysis reveals specific segmentation patterns observed in its absence. Finally, we find that the label set specificity does not affect dialog act segmentation performance. These findings have significant practical implications for spoken language understanding applications that depend heavily on a good-quality segmentation being available.
An intelligent virtual assistant (IVA) enables effortless conversations in call routing through spoken utterance classification (SUC) which is a special form of spoken language understanding (SLU). Building a SUC system requires a large amount of supervised in-domain data that is not always available. In this paper, we introduce an unsupervised spoken utterance classification approach (USUC) that does not require any in-domain data except for the intent labels and a few para-phrases per intent. USUC is consisting of a KNN classifier (K=1) and a complex embedding model trained on a large amount of unsupervised customer service corpus. Among all embedding models, we demonstrate that Elmo works best for USUC. However, an Elmo model is too slow to be used at run-time for call routing. To resolve this issue, first, we compute the uni- and bi-gram embedding vectors offline and we build a lookup table of n-grams and their corresponding embedding vector. Then we use this table to compute sentence embedding vectors at run-time, along with back-off techniques for unseen n-grams. Experiments show that USUC outperforms the traditional utterance classification methods by reducing the classification error rate from 32.9% to 27.0% without requiring supervised data. Moreover, our lookup and back-off technique increases the processing speed from 16 utterances per second to 118 utterances per second.
Spoken dialogue systems such as Siri and Alexa provide great convenience to people's everyday life. However, current spoken language understanding (SLU) pipelines largely depend on automatic speech recognition (ASR) modules, which require a large amount of language-specific training data. In this paper, we propose a Transformer-based SLU system that works directly on phones. This acoustic-based SLU system consists of only two blocks and does not require the presence of ASR module. The first block is a universal phone recognition system, and the second block is a Transformer-based language model for phones. We verify the effectiveness of the system on an intent classification dataset in Mandarin Chinese.
Spoken language understanding (SLU) systems extract both text transcripts and semantics associated with intents and slots from input speech utterances. SLU systems usually consist of (1) an automatic speech recognition (ASR) module, (2) an interface module that exposes relevant outputs from ASR, and (3) a natural language understanding (NLU) module. Interfaces in SLU systems carry information on text transcriptions or richer information like neural embeddings from ASR to NLU. In this paper, we study how interfaces affect joint-training for spoken language understanding. Most notably, we obtain the state-of-the-art results on the publicly available 50-hr SLURP dataset. We first leverage large-size pretrained ASR and NLU models that are connected by a text interface, and then jointly train both models via a sequence loss function. For scenarios where pretrained models are not utilized, the best results are obtained through a joint sequence loss training using richer neural interfaces. Finally, we show the overall diminishing impact of leveraging pretrained models with increased training data size.
Decomposable tasks are complex and comprise of a hierarchy of sub-tasks. Spoken intent prediction, for example, combines automatic speech recognition and natural language understanding. Existing benchmarks, however, typically hold out examples for only the surface-level sub-task. As a result, models with similar performance on these benchmarks may have unobserved performance differences on the other sub-tasks. To allow insightful comparisons between competitive end-to-end architectures, we propose a framework to construct robust test sets using coordinate ascent over sub-task specific utility functions. Given a dataset for a decomposable task, our method optimally creates a test set for each sub-task to individually assess sub-components of the end-to-end model. Using spoken language understanding as a case study, we generate new splits for the Fluent Speech Commands and Snips SmartLights datasets. Each split has two test sets: one with held-out utterances assessing natural language understanding abilities, and one with held-out speakers to test speech processing skills. Our splits identify performance gaps up to 10% between end-to-end systems that were within 1% of each other on the original test sets. These performance gaps allow more realistic and actionable comparisons between different architectures, driving future model development. We release our splits and tools for the community.
Most of the existing spoken language understanding systems can perform only semantic frame parsing based on a single-round user query. They cannot take users' feedback to update/add/remove slot values through multiround interactions with users. In this paper, we introduce a novel multi-step spoken language understanding system based on adversarial learning that can leverage the multiround user's feedback to update slot values. We perform two experiments on the benchmark ATIS dataset and demonstrate that the new system can improve parsing performance by at least $2.5\%$ in terms of F1, with only one round of feedback. The improvement becomes even larger when the number of feedback rounds increases. Furthermore, we also compare the new system with state-of-the-art dialogue state tracking systems and demonstrate that the new interactive system can perform better on multiround spoken language understanding tasks in terms of slot- and sentence-level accuracy.
Intent classification is a major task in spoken language understanding (SLU). Since most models are built with pre-collected in-domain (IND) training utterances, their ability to detect unsupported out-of-domain (OOD) utterances has a critical effect in practical use. Recent works have shown that using extra data and labels can improve the OOD detection performance, yet it could be costly to collect such data. This paper proposes to train a model with only IND data while supporting both IND intent classification and OOD detection. Our method designs a novel domain-regularized module (DRM) to reduce the overconfident phenomenon of a vanilla classifier, achieving a better generalization in both cases. Besides, DRM can be used as a drop-in replacement for the last layer in any neural network-based intent classifier, providing a low-cost strategy for a significant improvement. The evaluation on four datasets shows that our method built on BERT and RoBERTa models achieves state-of-the-art performance against existing approaches and the strong baselines we created for the comparisons.
Spoken language understanding (SLU) topic has seen a lot of progress these last three years, with the emergence of end-to-end neural approaches. Spoken language understanding refers to natural language processing tasks related to semantic extraction from speech signal, like named entity recognition from speech or slot filling task in a context of human-machine dialogue. Classically, SLU tasks were processed through a cascade approach that consists in applying, firstly, an automatic speech recognition process, followed by a natural language processing module applied to the automatic transcriptions. These three last years, end-to-end neural approaches, based on deep neural networks, have been proposed in order to directly extract the semantics from speech signal, by using a single neural model. More recent works on self-supervised training with unlabeled data open new perspectives in term of performance for automatic speech recognition and natural language processing. In this paper, we present a brief overview of the recent advances on the French MEDIA benchmark dataset for SLU, with or without the use of additional data. We also present our last results that significantly outperform the current state-of-the-art with a Concept Error Rate (CER) of 11.2%, instead of 13.6% for the last state-of-the-art system presented this year.
End-to-end (E2E) spoken language understanding (SLU) systems predict utterance semantics directly from speech using a single model. Previous work in this area has focused on targeted tasks in fixed domains, where the output semantic structure is assumed a priori and the input speech is of limited complexity. In this work we present our approach to developing an E2E model for generalized SLU in commercial voice assistants (VAs). We propose a fully differentiable, transformer-based, hierarchical system that can be pretrained at both the ASR and NLU levels. This is then fine-tuned on both transcription and semantic classification losses to handle a diverse set of intent and argument combinations. This leads to an SLU system that achieves significant improvements over baselines on a complex internal generalized VA dataset with a 43% improvement in accuracy, while still meeting the 99% accuracy benchmark on the popular Fluent Speech Commands dataset. We further evaluate our model on a hard test set, exclusively containing slot arguments unseen in training, and demonstrate a nearly 20% improvement, showing the efficacy of our approach in truly demanding VA scenarios.
End-to-end (E2E) spoken language understanding (SLU) systems avoid an intermediate textual representation by mapping speech directly into intents with slot values. This approach requires considerable domain-specific training data. In low-resource scenarios this is a major concern, e.g., in the present study dealing with SLU for dysarthric speech. Pretraining part of the SLU model for automatic speech recognition targets helps but no research has shown to which extent SLU on dysarthric speech benefits from knowledge transferred from other dysarthric speech tasks. This paper investigates the efficiency of pre-training strategies for SLU tasks on dysarthric speech. The designed SLU system consists of a TDNN acoustic model for feature encoding and a capsule network for intent and slot decoding. The acoustic model is pre-trained in two stages: initialization with a corpus of normal speech and finetuning on a mixture of dysarthric and normal speech. By introducing the intelligibility score as a metric of the impairment severity, this paper quantitatively analyzes the relation between generalization and pathology severity for dysarthric speech.
Spoken Language Understanding (SLU) systems parse speech into semantic structures like dialog acts and slots. This involves the use of an Automatic Speech Recognizer (ASR) to transcribe speech into multiple text alternatives (hypotheses). Transcription errors, common in ASRs, impact downstream SLU performance negatively. Approaches to mitigate such errors involve using richer information from the ASR, either in form of N-best hypotheses or word-lattices. We hypothesize that transformer models learn better with a simpler utterance representation using the concatenation of the N-best ASR alternatives, where each alternative is separated by a special delimiter [SEP]. In our work, we test our hypothesis by using concatenated N-best ASR alternatives as the input to transformer encoder models, namely BERT and XLM-RoBERTa, and achieve performance equivalent to the prior state-of-the-art model on DSTC2 dataset. We also show that our approach significantly outperforms the prior state-of-the-art when subjected to the low data regime. Additionally, this methodology is accessible to users of third-party ASR APIs which do not provide word-lattice information.
Human-computer interaction (HCI) is significantly impacted by delayed responses from a spoken dialogue system. Hence, end-to-end (e2e) spoken language understanding (SLU) solutions have recently been proposed to decrease latency. Such approaches allow for the extraction of semantic information directly from the speech signal, thus bypassing the need for a transcript from an automatic speech recognition (ASR) system. In this paper, we propose a compact e2e SLU architecture for streaming scenarios, where chunks of the speech signal are processed continuously to predict intent and slot values. Our model is based on a 3D convolutional neural network (3D-CNN) and a unidirectional long short-term memory (LSTM). We compare the performance of two alignment-free losses: the connectionist temporal classification (CTC) method and its adapted version, namely connectionist temporal localization (CTL). The latter performs not only the classification but also localization of sequential audio events. The proposed solution is evaluated on the Fluent Speech Command dataset and results show our model ability to process incoming speech signal, reaching accuracy as high as 98.97 % for CTC and 98.78 % for CTL on single-label classification, and as high as 95.69 % for CTC and 95.28 % for CTL on two-label prediction.
SpeechBrain is an open-source and all-in-one speech toolkit. It is designed to facilitate the research and development of neural speech processing technologies by being simple, flexible, user-friendly, and well-documented. This paper describes the core architecture designed to support several tasks of common interest, allowing users to naturally conceive, compare and share novel speech processing pipelines. SpeechBrain achieves competitive or state-of-the-art performance in a wide range of speech benchmarks. It also provides training recipes, pretrained models, and inference scripts for popular speech datasets, as well as tutorials which allow anyone with basic Python proficiency to familiarize themselves with speech technologies.
Recent advances in supervised, semi-supervised and self-supervised deep learning algorithms have shown significant improvement in the performance of automatic speech recognition(ASR) systems. The state-of-the-art systems have achieved a word error rate (WER) less than 5%. However, in the past, researchers have argued the non-suitability of the WER metric for the evaluation of ASR systems for downstream tasks such as spoken language understanding (SLU) and information retrieval. The reason is that the WER works at the surface level and does not include any syntactic and semantic knowledge.The current work proposes Semantic-WER (SWER), a metric to evaluate the ASR transcripts for downstream applications in general. The SWER can be easily customized for any down-stream task.
Named entity recognition (NER) is a fundamental component in many applications, such as Web Search and Voice Assistants. Although deep neural networks greatly improve the performance of NER, due to the requirement of large amounts of training data, deep neural networks can hardly scale out to many languages in an industry setting. To tackle this challenge, cross-lingual NER transfers knowledge from a rich-resource language to languages with low resources through pre-trained multilingual language models. Instead of using training data in target languages, cross-lingual NER has to rely on only training data in source languages, and optionally adds the translated training data derived from source languages. However, the existing cross-lingual NER methods do not make good use of rich unlabeled data in target languages, which is relatively easy to collect in industry applications. To address the opportunities and challenges, in this paper we describe our novel practice in Microsoft to leverage such large amounts of unlabeled data in target languages in real production settings. To effectively extract weak supervision signals from the unlabeled data, we develop a novel approach based on the ideas of semi-supervised learning and reinforcement learning. The empirical study on three benchmark data sets verifies that our approach establishes the new state-of-the-art performance with clear edges. Now, the NER techniques reported in this paper are on their way to become a fundamental component for Web ranking, Entity Pane, Answers Triggering, and Question Answering in the Microsoft Bing search engine. Moreover, our techniques will also serve as part of the Spoken Language Understanding module for a commercial voice assistant. We plan to open source the code of the prototype framework after deployment.
End-to-end spoken language understanding (SLU) has recently attracted increasing interest. Compared to the conventional tandem-based approach that combines speech recognition and language understanding as separate modules, the new approach extracts users' intentions directly from the speech signals, resulting in joint optimization and low latency. Such an approach, however, is typically designed to process one intention at a time, which leads users to take multiple rounds to fulfill their requirements while interacting with a dialogue system. In this paper, we propose a streaming end-to-end framework that can process multiple intentions in an online and incremental way. The backbone of our framework is a unidirectional RNN trained with the connectionist temporal classification (CTC) criterion. By this design, an intention can be identified when sufficient evidence has been accumulated, and multiple intentions can be identified sequentially. We evaluate our solution on the Fluent Speech Commands (FSC) dataset and the intent detection accuracy is about 97 % on all multi-intent settings. This result is comparable to the performance of the state-of-the-art non-streaming models, but is achieved in an online and incremental way. We also employ our model to a keyword spotting task using the Google Speech Commands dataset and the results are also highly promising.
The lack of publicly available evaluation data for low-resource languages limits progress in Spoken Language Understanding (SLU). As key tasks like intent classification and slot filling require abundant training data, it is desirable to reuse existing data in high-resource languages to develop models for low-resource scenarios. We introduce xSID, a new benchmark for cross-lingual Slot and Intent Detection in 13 languages from 6 language families, including a very low-resource dialect. To tackle the challenge, we propose a joint learning approach, with English SLU training data and non-English auxiliary tasks from raw text, syntax and translation for transfer. We study two setups which differ by type and language coverage of the pre-trained embeddings. Our results show that jointly learning the main tasks with masked language modeling is effective for slots, while machine translation transfer works best for intent classification.
Generalization is a central problem in machine learning, especially when data is limited. Using prior information to enforce constraints is the principled way of encouraging generalization. In this work, we propose to leverage the prior information embedded in pretrained language models (LM) to improve generalization for intent classification and slot labeling tasks with limited training data. Specifically, we extract prior knowledge from pretrained LM in the form of synthetic data, which encode the prior implicitly. We fine-tune the LM to generate an augmented language, which contains not only text but also encodes both intent labels and slot labels. The generated synthetic data can be used to train a classifier later. Since the generated data may contain noise, we rephrase the learning from generated data as learning with noisy labels. We then utilize the mixout regularization for the classifier and prove its effectiveness to resist label noise in generated data. Empirically, our method demonstrates superior performance and outperforms the baseline by a large margin.
In contrast to conventional pipeline Spoken Language Understanding (SLU) which consists of automatic speech recognition (ASR) and natural language understanding (NLU), end-to-end SLU infers the semantic meaning directly from speech and overcomes the error propagation caused by ASR. End-to-end slot filling (SF) from speech is an essential component of end-to-end SLU, and is usually regarded as a sequence-to-sequence generation problem, heavily relied on the performance of language model of ASR. However, it is hard to generate a correct slot when the slot is out-of-vovabulary (OOV) in training data, especially when a slot is an anti-linguistic entity without grammatical rule. Inspired by object detection in computer vision that is to detect the object from an image, we consider SF as the task of slot detection from speech. In this paper, we formulate the SF task as a matching task and propose an end-to-end knowledge-based SF model, named Speech-to-Slot (Speech2Slot), to leverage knowledge to detect the boundary of a slot from the speech. We also release a large-scale dataset of Chinese speech for slot filling, containing more than 830,000 samples. The experiments show that our approach is markedly superior to the conventional pipeline SLU approach, and outperforms the state-of-the-art end-to-end SF approach with 12.51% accuracy improvement.