2023-07-20 更新
Leveraging Pretrained ASR Encoders for Effective and Efficient End-to-End Speech Intent Classification and Slot Filling
Authors:He Huang, Jagadeesh Balam, Boris Ginsburg
We study speech intent classification and slot filling (SICSF) by proposing to use an encoder pretrained on speech recognition (ASR) to initialize an end-to-end (E2E) Conformer-Transformer model, which achieves the new state-of-the-art results on the SLURP dataset, with 90.14% intent accuracy and 82.27% SLURP-F1. We compare our model with encoders pretrained on self-supervised learning (SSL), and show that ASR pretraining is much more effective than SSL for SICSF. To explore parameter efficiency, we freeze the encoder and add Adapter modules, and show that parameter efficiency is only achievable with an ASR-pretrained encoder, while the SSL encoder needs full finetuning to achieve comparable results. In addition, we provide an in-depth comparison on end-to-end models versus cascading models (ASR+NLU), and show that E2E models are better than cascaded models unless an oracle ASR model is provided. Last but not least, our model is the first E2E model that achieves the same performance as cascading models with oracle ASR. Code, checkpoints and configs are available.
PDF INTERSPEECH 2023
点此查看论文截图
Representation Learning With Hidden Unit Clustering For Low Resource Speech Applications
Authors:Varun Krishna, Tarun Sai, Sriram Ganapathy
The representation learning of speech, without textual resources, is an area of significant interest for many low resource speech applications. In this paper, we describe an approach to self-supervised representation learning from raw audio using a hidden unit clustering (HUC) framework. The input to the model consists of audio samples that are windowed and processed with 1-D convolutional layers. The learned “time-frequency” representations from the convolutional neural network (CNN) module are further processed with long short term memory (LSTM) layers which generate a contextual vector representation for every windowed segment. The HUC framework, allowing the categorization of the representations into a small number of phoneme-like units, is used to train the model for learning semantically rich speech representations. The targets consist of phoneme-like pseudo labels for each audio segment and these are generated with an iterative k-means algorithm. We explore techniques that improve the speaker invariance of the learned representations and illustrate the effectiveness of the proposed approach on two settings, i) completely unsupervised speech applications on the sub-tasks described as part of the ZeroSpeech 2021 challenge and ii) semi-supervised automatic speech recognition (ASR) applications on the TIMIT dataset and on the GramVaani challenge Hindi dataset. In these experiments, we achieve state-of-art results for various ZeroSpeech tasks. Further, on the ASR experiments, the HUC representations are shown to improve significantly over other established benchmarks based on Wav2vec, HuBERT and Best-RQ.
PDF
点此查看论文截图
Sumformer: A Linear-Complexity Alternative to Self-Attention for Speech Recognition
Authors:Titouan Parcollet, Rogier van Dalen, Shucong Zhang, Sourav Bhattacharya
Modern speech recognition systems rely on self-attention. Unfortunately, token mixing with self-attention takes quadratic time in the length of the speech utterance, slowing down inference as well as training and increasing memory consumption. Cheaper alternatives to self-attention for ASR have been developed, but fail to consistently reach the same level of accuracy. In practice, however, the self-attention weights of trained speech recognizers take the form of a global average over time. This paper, therefore, proposes a linear-time alternative to self-attention for speech recognition. It summarises a whole utterance with the mean over vectors for all time steps. This single summary is then combined with time-specific information. We call this method ``Summary Mixing’’. Introducing Summary Mixing in state-of-the-art ASR models makes it feasible to preserve or exceed previous speech recognition performance while lowering the training and inference times by up to 27% and reducing the memory budget by a factor of two.
PDF Submitted to NeurIPS 2023
点此查看论文截图
Towards spoken dialect identification of Irish
Authors:Liam Lonergan, Mengjie Qian, Neasa Ní Chiaráin, Christer Gobl, Ailbhe Ní Chasaide
The Irish language is rich in its diversity of dialects and accents. This compounds the difficulty of creating a speech recognition system for the low-resource language, as such a system must contend with a high degree of variability with limited corpora. A recent study investigating dialect bias in Irish ASR found that balanced training corpora gave rise to unequal dialect performance, with performance for the Ulster dialect being consistently worse than for the Connacht or Munster dialects. Motivated by this, the present experiments investigate spoken dialect identification of Irish, with a view to incorporating such a system into the speech recognition pipeline. Two acoustic classification models are tested, XLS-R and ECAPA-TDNN, in conjunction with a text-based classifier using a pretrained Irish-language BERT model. The ECAPA-TDNN, particularly a model pretrained for language identification on the VoxLingua107 dataset, performed best overall, with an accuracy of 73%. This was further improved to 76% by fusing the model’s outputs with the text-based model. The Ulster dialect was most accurately identified, with an accuracy of 94%, however the model struggled to disambiguate between the Connacht and Munster dialects, suggesting a more nuanced approach may be necessary to robustly distinguish between the dialects of Irish.
PDF Accepted to Interspeech 2023 Workshop of the 2nd Annual Meeting of the Special Interest Group of Under-resourced Languages Workshop, Dublin (SiGUL)
点此查看论文截图
Audio-Visual Speech Enhancement Using Self-supervised Learning to Improve Speech Intelligibility in Cochlear Implant Simulations
Authors:Richard Lee Lai, Jen-Cheng Hou, Mandar Gogate, Kia Dashtipour, Amir Hussain, Yu Tsao
Individuals with hearing impairments face challenges in their ability to comprehend speech, particularly in noisy environments. The aim of this study is to explore the effectiveness of audio-visual speech enhancement (AVSE) in enhancing the intelligibility of vocoded speech in cochlear implant (CI) simulations. Notably, the study focuses on a challenged scenario where there is limited availability of training data for the AVSE task. To address this problem, we propose a novel deep neural network framework termed Self-Supervised Learning-based AVSE (SSL-AVSE). The proposed SSL-AVSE combines visual cues, such as lip and mouth movements, from the target speakers with corresponding audio signals. The contextually combined audio and visual data are then fed into a Transformer-based SSL AV-HuBERT model to extract features, which are further processed using a BLSTM-based SE model. The results demonstrate several key findings. Firstly, SSL-AVSE successfully overcomes the issue of limited data by leveraging the AV-HuBERT model. Secondly, by fine-tuning the AV-HuBERT model parameters for the target SE task, significant performance improvements are achieved. Specifically, there is a notable enhancement in PESQ (Perceptual Evaluation of Speech Quality) from 1.43 to 1.67 and in STOI (Short-Time Objective Intelligibility) from 0.70 to 0.74. Furthermore, the performance of the SSL-AVSE was evaluated using CI vocoded speech to assess the intelligibility for CI users. Comparative experimental outcomes reveal that in the presence of dynamic noises encountered during human conversations, SSL-AVSE exhibits a substantial improvement. The NCM (Normal Correlation Matrix) values indicate an increase of 26.5% to 87.2% compared to the noisy baseline.
PDF
点此查看论文截图
Efficient Region-Aware Neural Radiance Fields for High-Fidelity Talking Portrait Synthesis
Authors:Jiahe Li, Jiawei Zhang, Xiao Bai, Jun Zhou, Lin Gu
This paper presents ER-NeRF, a novel conditional Neural Radiance Fields (NeRF) based architecture for talking portrait synthesis that can concurrently achieve fast convergence, real-time rendering, and state-of-the-art performance with small model size. Our idea is to explicitly exploit the unequal contribution of spatial regions to guide talking portrait modeling. Specifically, to improve the accuracy of dynamic head reconstruction, a compact and expressive NeRF-based Tri-Plane Hash Representation is introduced by pruning empty spatial regions with three planar hash encoders. For speech audio, we propose a Region Attention Module to generate region-aware condition feature via an attention mechanism. Different from existing methods that utilize an MLP-based encoder to learn the cross-modal relation implicitly, the attention mechanism builds an explicit connection between audio features and spatial regions to capture the priors of local motions. Moreover, a direct and fast Adaptive Pose Encoding is introduced to optimize the head-torso separation problem by mapping the complex transformation of the head pose into spatial coordinates. Extensive experiments demonstrate that our method renders better high-fidelity and audio-lips synchronized talking portrait videos, with realistic details and high efficiency compared to previous methods.
PDF Accepted by ICCV 2023
点此查看论文截图
SLMGAN: Exploiting Speech Language Model Representations for Unsupervised Zero-Shot Voice Conversion in GANs
Authors:Yinghao Aaron Li, Cong Han, Nima Mesgarani
In recent years, large-scale pre-trained speech language models (SLMs) have demonstrated remarkable advancements in various generative speech modeling applications, such as text-to-speech synthesis, voice conversion, and speech enhancement. These applications typically involve mapping text or speech inputs to pre-trained SLM representations, from which target speech is decoded. This paper introduces a new approach, SLMGAN, to leverage SLM representations for discriminative tasks within the generative adversarial network (GAN) framework, specifically for voice conversion. Building upon StarGANv2-VC, we add our novel SLM-based WavLM discriminators on top of the mel-based discriminators along with our newly designed SLM feature matching loss function, resulting in an unsupervised zero-shot voice conversion system that does not require text labels during training. Subjective evaluation results show that SLMGAN outperforms existing state-of-the-art zero-shot voice conversion models in terms of naturalness and achieves comparable similarity, highlighting the potential of SLM-based discriminators for related applications.
PDF WASPAA 2023
点此查看论文截图
Enhancing conversational quality in language learning chatbots: An evaluation of GPT4 for ASR error correction
Authors:Long Mai, Julie Carson-Berndsen
The integration of natural language processing (NLP) technologies into educational applications has shown promising results, particularly in the language learning domain. Recently, many spoken open-domain chatbots have been used as speaking partners, helping language learners improve their language skills. However, one of the significant challenges is the high word-error-rate (WER) when recognizing non-native/non-fluent speech, which interrupts conversation flow and leads to disappointment for learners. This paper explores the use of GPT4 for ASR error correction in conversational settings. In addition to WER, we propose to use semantic textual similarity (STS) and next response sensibility (NRS) metrics to evaluate the impact of error correction models on the quality of the conversation. We find that transcriptions corrected by GPT4 lead to higher conversation quality, despite an increase in WER. GPT4 also outperforms standard error correction methods without the need for in-domain training data.
PDF
点此查看论文截图
Leveraging Visemes for Better Visual Speech Representation and Lip Reading
Authors:Javad Peymanfard, Vahid Saeedi, Mohammad Reza Mohammadi, Hossein Zeinali, Nasser Mozayani
Lip reading is a challenging task that has many potential applications in speech recognition, human-computer interaction, and security systems. However, existing lip reading systems often suffer from low accuracy due to the limitations of video features. In this paper, we propose a novel approach that leverages visemes, which are groups of phonetically similar lip shapes, to extract more discriminative and robust video features for lip reading. We evaluate our approach on various tasks, including word-level and sentence-level lip reading, and audiovisual speech recognition using the Arman-AV dataset, a largescale Persian corpus. Our experimental results show that our viseme based approach consistently outperforms the state-of-theart methods in all these tasks. The proposed method reduces the lip-reading word error rate (WER) by 9.1% relative to the best previous method.
PDF
点此查看论文截图