Search

Continuous speech can be converted into a discrete sequence by deriving discrete units from the hidden features of self-supervised learned (SSL) speech models. Although SSL models are becoming larger and trained on more data, they are often sensitive to real-life distortions like additive noise or reverberation, which translates to a shift in discrete units. We propose a parameter-efficient approach to generate noise-robust discrete units from pre-trained SSL models by training a small encoder-decoder model, with or without adapters, to simultaneously denoise and discretise the hidden features of the SSL model. The model learns to generate a clean discrete sequence for a noisy utterance, conditioned on the SSL features. The proposed denoiser outperforms several pre-training methods on the tasks of noisy discretisation and noisy speech recognition, and can be finetuned to the target environment with a few recordings of unlabeled target data., Comment: Accepted at SLT2024

Automated audio captioning (AAC) is an audio-to-text task to describe audio contents in natural language. Recently, the advancements in large language models (LLMs), with improvements in training approaches for audio encoders, have opened up possibilities for improving AAC. Thus, we explore enhancing AAC from three aspects: 1) a pre-trained audio encoder via consistent ensemble distillation (CED) is used to improve the effectivity of acoustic tokens, with a querying transformer (Q-Former) bridging the modality gap to LLM and compress acoustic tokens; 2) we investigate the advantages of using a Llama 2 with 7B parameters as the decoder; 3) another pre-trained LLM corrects text errors caused by insufficient training data and annotation ambiguities. Both the audio encoder and text decoder are optimized by low-rank adaptation (LoRA). Experiments show that each of these enhancements is effective. Our method obtains a 33.0 SPIDEr-FL score, outperforming the winner of DCASE 2023 Task 6A., Comment: Accepted by Interspeech 2024

Audio-text retrieval is a challenging task, requiring the search for an audio clip or a text caption within a database. The predominant focus of existing research on English descriptions poses a limitation on the applicability of such models, given the abundance of non-English content in real-world data. To address these linguistic disparities, we propose a language enhancement (LE), using a multilingual text encoder (SONAR) to encode the text data with language-specific information. Additionally, we optimize the audio encoder through the application of consistent ensemble distillation (CED), enhancing support for variable-length audio-text retrieval. Our methodology excels in English audio-text retrieval, demonstrating state-of-the-art (SOTA) performance on commonly used datasets such as AudioCaps and Clotho. Simultaneously, the approach exhibits proficiency in retrieving content in seven other languages with only 10% of additional language-enhanced training data, yielding promising results. The source code is publicly available https://github.com/zyyan4/ml-clap., Comment: interspeech2024

Despite progress in audio classification, a generalization gap remains between speech and other sound domains, such as environmental sounds and music. Models trained for speech tasks often fail to perform well on environmental or musical audio tasks, and vice versa. While self-supervised (SSL) audio representations offer an alternative, there has been limited exploration of scaling both model and dataset sizes for SSL-based general audio classification. We introduce Dasheng, a simple SSL audio encoder, based on the efficient masked autoencoder framework. Trained with 1.2 billion parameters on 272,356 hours of diverse audio, Dasheng obtains significant performance gains on the HEAR benchmark. It outperforms previous works on CREMA-D, LibriCount, Speech Commands, VoxLingua, and competes well in music and environment classification. Dasheng features inherently contain rich speech, music, and environmental information, as shown in nearest-neighbor classification experiments. Code is available https://github.com/richermans/dasheng/., Comment: Interspeech 2024

Recent research in zero-shot speech synthesis has made significant progress in speaker similarity. However, current efforts focus on timbre generalization rather than prosody modeling, which results in limited naturalness and expressiveness. To address this, we introduce a novel speech synthesis model trained on large-scale datasets, including both timbre and hierarchical prosody modeling. As timbre is a global attribute closely linked to expressiveness, we adopt a global vector to model speaker timbre while guiding prosody modeling. Besides, given that prosody contains both global consistency and local variations, we introduce a diffusion model as the pitch predictor and employ a prosody adaptor to model prosody hierarchically, further enhancing the prosody quality of the synthesized speech. Experimental results show that our model not only maintains comparable timbre quality to the baseline but also exhibits better naturalness and expressiveness., Comment: 5 pages, 2 figures, accepted by Interspeech2024

Augmentation and knowledge distillation (KD) are well-established techniques employed in audio classification tasks, aimed at enhancing performance and reducing model sizes on the widely recognized Audioset (AS) benchmark. Although both techniques are effective individually, their combined use, called consistent teaching, hasn't been explored before. This paper proposes CED, a simple training framework that distils student models from large teacher ensembles with consistent teaching. To achieve this, CED efficiently stores logits as well as the augmentation methods on disk, making it scalable to large-scale datasets. Central to CED's efficacy is its label-free nature, meaning that only the stored logits are used for the optimization of a student model only requiring 0.3\% additional disk space for AS. The study trains various transformer-based models, including a 10M parameter model achieving a 49.0 mean average precision (mAP) on AS. Pretrained models and code are available at https://github.com/RicherMans/CED.

Recently, deep learning-based beamforming algorithms have shown promising performance in target speech extraction tasks. However, most systems do not fully utilize spatial information. In this paper, we propose a target speech extraction network that utilizes spatial information to enhance the performance of neural beamformer. To achieve this, we first use the UNet-TCN structure to model input features and improve the estimation accuracy of the speech pre-separation module by avoiding information loss caused by direct dimensionality reduction in other models. Furthermore, we introduce a multi-head cross-attention mechanism that enhances the neural beamformer's perception of spatial information by making full use of the spatial information received by the array. Experimental results demonstrate that our approach, which incorporates a more reasonable target mask estimation network and a spatial information-based cross-attention mechanism into the neural beamformer, effectively improves speech separation performance.

Previously, Target Speaker Extraction (TSE) has yielded outstanding performance in certain application scenarios for speech enhancement and source separation. However, obtaining auxiliary speaker-related information is still challenging in noisy environments with significant reverberation. inspired by the recently proposed distance-based sound separation, we propose the near sound (NS) extractor, which leverages distance information for TSE to reliably extract speaker information without requiring previous speaker enrolment, called speaker embedding self-enrollment (SESE). Full- & sub-band modeling is introduced to enhance our NS-Extractor's adaptability towards environments with significant reverberation. Experimental results on several cross-datasets demonstrate the effectiveness of our improvements and the excellent performance of our proposed NS-Extractor in different application scenarios., Comment: Proc. INTERSPEECH 2023, 2488-2492, doi: 10.21437/Interspeech.2023-218

Visual information can serve as an effective cue for target speaker extraction (TSE) and is vital to improving extraction performance. In this paper, we propose AV-SepFormer, a SepFormer-based attention dual-scale model that utilizes cross- and self-attention to fuse and model features from audio and visual. AV-SepFormer splits the audio feature into a number of chunks, equivalent to the length of the visual feature. Then self- and cross-attention are employed to model and fuse the multi-modal features. Furthermore, we use a novel 2D positional encoding, that introduces the positional information between and within chunks and provides significant gains over the traditional positional encoding. Our model has two key advantages: the time granularity of audio chunked feature is synchronized to the visual feature, which alleviates the harm caused by the inconsistency of audio and video sampling rate; by combining self- and cross-attention, feature fusion and speech extraction processes are unified within an attention paradigm. The experimental results show that AV-SepFormer significantly outperforms other existing methods., Comment: Accepted by ICASSP2023

The currently most prominent algorithm to train keyword spotting (KWS) models with deep neural networks (DNNs) requires strong supervision i.e., precise knowledge of the spoken keyword location in time. Thus, most KWS approaches treat the presence of redundant data, such as noise, within their training set as an obstacle. A common training paradigm to deal with data redundancies is to use temporally weakly supervised learning, which only requires providing labels on a coarse scale. This study explores the limits of DNN training using temporally weak labeling with applications in KWS. We train a simple end-to-end classifier on the common Google Speech Commands dataset with increased difficulty by randomly appending and adding noise to the training dataset. Our results indicate that temporally weak labeling can achieve comparable results to strongly supervised baselines while having a less stringent labeling requirement. In the presence of noise, weakly supervised models are capable to localize and extract target keywords without explicit supervision, leading to a performance increase compared to strongly supervised approaches.

Transformers have emerged as a prominent model framework for audio tagging (AT), boasting state-of-the-art (SOTA) performance on the widely-used Audioset dataset. However, their impressive performance often comes at the cost of high memory usage, slow inference speed, and considerable model delay, rendering them impractical for real-world AT applications. In this study, we introduce streaming audio transformers (SAT) that combine the vision transformer (ViT) architecture with Transformer-Xl-like chunk processing, enabling efficient processing of long-range audio signals. Our proposed SAT is benchmarked against other transformer-based SOTA methods, achieving significant improvements in terms of mean average precision (mAP) at a delay of 2s and 1s, while also exhibiting significantly lower memory usage and computational overhead. Checkpoints are publicly available https://github.com/RicherMans/SAT., Comment: Interspeech2024

In this paper, we investigate representation learning for low-resource keyword spotting (KWS). The main challenges of KWS are limited labeled data and limited available device resources. To address those challenges, we explore representation learning for KWS by self-supervised contrastive learning and self-training with pretrained model. First, local-global contrastive siamese networks (LGCSiam) are designed to learn similar utterance-level representations for similar audio samplers by proposed local-global contrastive loss without requiring ground-truth. Second, a self-supervised pretrained Wav2Vec 2.0 model is applied as a constraint module (WVC) to force the KWS model to learn frame-level acoustic representations. By the LGCSiam and WVC modules, the proposed small-footprint KWS model can be pretrained with unlabeled data. Experiments on speech commands dataset show that the self-training WVC module and the self-supervised LGCSiam module significantly improve accuracy, especially in the case of training on a small labeled dataset.

Electroencephalography (EEG) plays a vital role in detecting how brain responses to different stimulus. In this paper, we propose a novel Shallow-Deep Attention-based Network (SDANet) to classify the correct auditory stimulus evoking the EEG signal. It adopts the Attention-based Correlation Module (ACM) to discover the connection between auditory speech and EEG from global aspect, and the Shallow-Deep Similarity Classification Module (SDSCM) to decide the classification result via the embeddings learned from the shallow and deep layers. Moreover, various training strategies and data augmentation are used to boost the model robustness. Experiments are conducted on the dataset provided by Auditory EEG challenge (ICASSP Signal Processing Grand Challenge 2023). Results show that the proposed model has a significant gain over the baseline on the match-mismatch track.

Existing weakly supervised sound event detection (WSSED) work has not explored both types of co-occurrences simultaneously, i.e., some sound events often co-occur, and their occurrences are usually accompanied by specific background sounds, so they would be inevitably entangled, causing misclassification and biased localization results with only clip-level supervision. To tackle this issue, we first establish a structural causal model (SCM) to reveal that the context is the main cause of co-occurrence confounders that mislead the model to learn spurious correlations between frames and clip-level labels. Based on the causal analysis, we propose a causal intervention (CI) method for WSSED to remove the negative impact of co-occurrence confounders by iteratively accumulating every possible context of each class and then re-projecting the contexts to the frame-level features for making the event boundary clearer. Experiments show that our method effectively improves the performance on multiple datasets and can generalize to various baseline models., Comment: Accepted by ICASSP2023

Keyword spotting (KWS) is a core human-machine-interaction front-end task for most modern intelligent assistants. Recently, a unified (UniKW-AT) framework has been proposed that adds additional capabilities in the form of audio tagging (AT) to a KWS model. However, previous work did not consider the real-world deployment of a UniKW-AT model, where factors such as model size and inference speed are more important than performance alone. This work introduces three mobile-device deployable models named Unified Transformers (UiT). Our best model achieves an mAP of 34.09 on Audioset, and an accuracy of 97.76 on the public Google Speech Commands V1 dataset. Further, we benchmark our proposed approaches on four mobile platforms, revealing that the proposed UiT models can achieve a speedup of 2 - 6 times against a competitive MobileNetV2., Comment: ICASSP 2023

In most cases, bilingual TTS needs to handle three types of input scripts: first language only, second language only, and second language embedded in the first language. In the latter two situations, the pronunciation and intonation of the second language are usually quite different due to the influence of the first language. Therefore, it is a big challenge to accurately model the pronunciation and intonation of the second language in different contexts without mutual interference. This paper builds a Mandarin-English TTS system to acquire more standard spoken English speech from a monolingual Chinese speaker. We introduce phonology embedding to capture the English differences between different phonology. Embedding mask is applied to language embedding for distinguishing information between different languages and to phonology embedding for focusing on English expression. We specially design an embedding strength modulator to capture the dynamic strength of language and phonology. Experiments show that our approach can produce significantly more natural and standard spoken English speech of the monolingual Chinese speaker. From analysis, we find that suitable phonology control contributes to better performance in different scenarios., Comment: Submitted to ICASSP2023

We study the usability of pre-trained weakly supervised audio tagging (AT) models as feature extractors for general audio representations. We mainly analyze the feasibility of transferring those embeddings to other tasks within the speech and sound domains. Specifically, we benchmark weakly supervised pre-trained models (MobileNetV2 and EfficientNet-B0) against modern self-supervised learning methods (BYOL-A) as feature extractors. Fourteen downstream tasks are used for evaluation ranging from music instrument classification to language classification. Our results indicate that AT pre-trained models are an excellent transfer learning choice for music, event, and emotion recognition tasks. Further, finetuning AT models can also benefit speech-related tasks such as keyword spotting and intent classification., Comment: Odyssey 2022

Within the audio research community and the industry, keyword spotting (KWS) and audio tagging (AT) are seen as two distinct tasks and research fields. However, from a technical point of view, both of these tasks are identical: they predict a label (keyword in KWS, sound event in AT) for some fixed-sized input audio segment. This work proposes UniKW-AT: An initial approach for jointly training both KWS and AT. UniKW-AT enhances the noise-robustness for KWS, while also being able to predict specific sound events and enabling conditional wake-ups on sound events. Our approach extends the AT pipeline with additional labels describing the presence of a keyword. Experiments are conducted on the Google Speech Commands V1 (GSCV1) and the balanced Audioset (AS) datasets. The proposed MobileNetV2 model achieves an accuracy of 97.53% on the GSCV1 dataset and an mAP of 33.4 on the AS evaluation set. Further, we show that significant noise-robustness gains can be observed on a real-world KWS dataset, greatly outperforming standard KWS approaches. Our study shows that KWS and AT can be merged into a single framework without significant performance degradation., Comment: Accepted in Interspeech2022

Large-scale audio tagging datasets inevitably contain imperfect labels, such as clip-wise annotated (temporally weak) tags with no exact on- and offsets, due to a high manual labeling cost. This work proposes pseudo strong labels (PSL), a simple label augmentation framework that enhances the supervision quality for large-scale weakly supervised audio tagging. A machine annotator is first trained on a large weakly supervised dataset, which then provides finer supervision for a student model. Using PSL we achieve an mAP of 35.95 balanced train subset of Audioset using a MobileNetV2 back-end, significantly outperforming approaches without PSL. An analysis is provided which reveals that PSL mitigates missing labels. Lastly, we show that models trained with PSL are also superior at generalizing to the Freesound datasets (FSD) than their weakly trained counterparts., Comment: Accepted by ICASSP 2022

Keyword spotting (KWS) and speaker verification (SV) are two important tasks in speech applications. Research shows that the state-of-art KWS and SV models are trained independently using different datasets since they expect to learn distinctive acoustic features. However, humans can distinguish language content and the speaker identity simultaneously. Motivated by this, we believe it is important to explore a method that can effectively extract common features while decoupling task-specific features. Bearing this in mind, a two-branch deep network (KWS branch and SV branch) with the same network structure is developed and a novel decoupling feature learning method is proposed to push up the performance of KWS and SV simultaneously where speaker-invariant keyword representations and keyword-invariant speaker representations are expected respectively. Experiments are conducted on Google Speech Commands Dataset (GSCD). The results demonstrate that the orthogonality regularization helps the network to achieve SOTA EER of 1.31% and 1.87% on KWS and SV, respectively., Comment: Accepted at ICASSP 2022

Human beings can perceive a target sound type from a multi-source mixture signal by the selective auditory attention, however, such functionality was hardly ever explored in machine hearing. This paper addresses the target sound detection (TSD) task, which aims to detect the target sound signal from a mixture audio when a target sound's reference audio is given. We present a novel target sound detection network (TSDNet) which consists of two main parts: A conditional network which aims at generating a sound-discriminative conditional embedding vector representing the target sound, and a detection network which takes both the mixture audio and the conditional embedding vector as inputs and produces the detection result of the target sound. These two networks can be jointly optimized with a multi-task learning approach to further improve the performance. In addition, we study both strong-supervised and weakly-supervised strategies to train TSDNet and propose a data augmentation method by mixing two samples. To facilitate this research, we build a target sound detection dataset (\textit{i.e.} URBAN-TSD) based on URBAN-SED and UrbanSound8K datasets, and experimental results indicate our method could get the segment-based F scores of 76.3$\%$ and 56.8$\%$ on the strongly-labelled and weakly-labelled data respectively., Comment: Submitted to DCASE workshop2022

Learning emotion embedding from reference audio is a straightforward approach for multi-emotion speech synthesis in encoder-decoder systems. But how to get better emotion embedding and how to inject it into TTS acoustic model more effectively are still under investigation. In this paper, we propose an innovative constraint to help VAE extract emotion embedding with better cluster cohesion. Besides, the obtained emotion embedding is used as query to aggregate latent representations of all encoder layers via attention. Moreover, the queries from encoder layers themselves are also helpful. Experiments prove the proposed methods can enhance the encoding of comprehensive syntactic and semantic information and produce more expressive emotional speech., Comment: accepted by ICASSP2022

Keyword spotting (KWS) on mobile devices generally requires a small memory footprint. However, most current models still maintain a large number of parameters in order to ensure good performance. To solve this problem, this paper proposes a separable temporal convolution neural network with attention, it has a small number of parameters. Through the time convolution combined with attention mechanism, a small number of parameters model (32.2K) is implemented while maintaining high performance. The proposed model achieves 95.7% accuracy on the Google Speech Commands dataset, which is close to the performance of Res15(239K), the state-of-the-art model in KWS at present., Comment: arXiv admin note: text overlap with arXiv:2108.12146

Keyword spotting (KWS) on mobile devices generally requires a small memory footprint. However, most current models still maintain a large number of parameters in order to ensure good performance. In this paper, we propose a temporally pooled attention module which can capture global features better than the AveragePool. Besides, we design a separable temporal convolution network which leverages depthwise separable and temporal convolution to reduce the number of parameter and calculations. Finally, taking advantage of separable temporal convolution and temporally pooled attention, a efficient neural network (ST-AttNet) is designed for KWS system. We evaluate the models on the publicly available Google speech commands data sets V1. The number of parameters of proposed model (48K) is 1/6 of state-of-the-art TC-ResNet14-1.5 model (305K). The proposed model achieves a 96.6% accuracy, which is comparable to the TC-ResNet14-1.5 model (96.6%).

We propose a multi-channel speech enhancement approach with a novel two-stage feature fusion method and a pre-trained acoustic model in a multi-task learning paradigm. In the first fusion stage, the time-domain and frequency-domain features are extracted separately. In the time domain, the multi-channel convolution sum (MCS) and the inter-channel convolution differences (ICDs) features are computed and then integrated with the first 2-D convolutional layer, while in the frequency domain, the log-power spectra (LPS) features from both original channels and super-directive beamforming outputs are combined with a second 2-D convolutional layer. To fully integrate the rich information of multi-channel speech, i.e. time-frequency domain features and the array geometry, we apply a third 2-D convolutional layer in the second fusion stage to obtain the final convolutional features. Furthermore, we propose to use a fixed clean acoustic model trained with the end-to-end lattice-free maximum mutual information criterion to enforce the enhanced output to have the same distribution as the clean waveform to alleviate the over-estimation problem of the enhancement task and constrain distortion. On the Task1 development dataset of ConferencingSpeech 2021 challenge, a PESQ improvement of 0.24 and 0.19 is attained compared to the official baseline and a recently proposed multi-channel separation method., Comment: 7 pages, 3 figures, accepted to APSIPA 2021, revised

In multi-speaker speech synthesis, data from a number of speakers usually tend to have great diversity due to the fact that the speakers may differ largely in ages, speaking styles, emotions, and so on. It is important but challenging to improve the modeling capabilities for multi-speaker speech synthesis. To address the issue, this paper proposes a high-capability speech synthesis system, called Msdtron, in which 1) a representation of the harmonic structure of speech, called excitation spectrogram, is designed to directly guide the learning of harmonics in mel-spectrogram. 2) conditional gated LSTM (CGLSTM) is proposed to control the flow of text content information through the network by re-weighting the gates of LSTM using speaker information. The experiments show a significant reduction in reconstruction error of mel-spectrogram in the training of the multi-speaker model, and a great improvement is observed in the subjective evaluation of speaker adapted model., Comment: Accepted by ICASSP-2022

This paper introduces GigaSpeech, an evolving, multi-domain English speech recognition corpus with 10,000 hours of high quality labeled audio suitable for supervised training, and 40,000 hours of total audio suitable for semi-supervised and unsupervised training. Around 40,000 hours of transcribed audio is first collected from audiobooks, podcasts and YouTube, covering both read and spontaneous speaking styles, and a variety of topics, such as arts, science, sports, etc. A new forced alignment and segmentation pipeline is proposed to create sentence segments suitable for speech recognition training, and to filter out segments with low-quality transcription. For system training, GigaSpeech provides five subsets of different sizes, 10h, 250h, 1000h, 2500h, and 10000h. For our 10,000-hour XL training subset, we cap the word error rate at 4% during the filtering/validation stage, and for all our other smaller training subsets, we cap it at 0%. The DEV and TEST evaluation sets, on the other hand, are re-processed by professional human transcribers to ensure high transcription quality. Baseline systems are provided for popular speech recognition toolkits, namely Athena, ESPnet, Kaldi and Pika.

This paper introduces a new open-source speech corpus named "speechocean762" designed for pronunciation assessment use, consisting of 5000 English utterances from 250 non-native speakers, where half of the speakers are children. Five experts annotated each of the utterances at sentence-level, word-level and phoneme-level. A baseline system is released in open source to illustrate the phoneme-level pronunciation assessment workflow on this corpus. This corpus is allowed to be used freely for commercial and non-commercial purposes. It is available for free download from OpenSLR, and the corresponding baseline system is published in the Kaldi speech recognition toolkit., Comment: Accepted in INTERSPEECH 2021

The front-end module in multi-channel automatic speech recognition (ASR) systems mainly use microphone array techniques to produce enhanced signals in noisy conditions with reverberation and echos. Recently, neural network (NN) based front-end has shown promising improvement over the conventional signal processing methods. In this paper, we propose to adopt the architecture of deep complex Unet (DCUnet) - a powerful complex-valued Unet-structured speech enhancement model - as the front-end of the multi-channel acoustic model, and integrate them in a multi-task learning (MTL) framework along with cascaded framework for comparison. Meanwhile, we investigate the proposed methods with several training strategies to improve the recognition accuracy on the 1000-hours real-world XiaoMi smart speaker data with echos. Experiments show that our proposed DCUnet-MTL method brings about 12.2% relative character error rate (CER) reduction compared with the traditional approach with array processing plus single-channel acoustic model. It also achieves superior performance than the recently proposed neural beamforming method., Comment: 7 pages, 4 figures, IEEE SLT 2021 Technical Committee

This paper presents the "Ethiopian" system for the SLT 2021 Children Speech Recognition Challenge. Various data processing and augmentation techniques are proposed to tackle children's speech recognition problem, especially the lack of the children's speech recognition training data issue. Detailed experiments are designed and conducted to show the effectiveness of each technique, across different speech recognition toolkits and model architectures. Step by step, we explain how we come up with our final system, which provides the state-of-the-art results in the SLT 2021 Children Speech Recognition Challenge, with 21.66% CER on the Track 1 evaluation set (4th place overall), and 16.53% CER on the Track 2 evaluation set (1st place overall). Post-challenge analysis shows that our system actually achieves 18.82% CER on the Track 1 evaluation set, but we submitted the wrong version to the challenge organizer for Track 1., Comment: System description of the SLT 2021 Children Speech Recognition Challenge

Smart audio devices are gated by an always-on lightweight keyword spotting program to reduce power consumption. It is however challenging to design models that have both high accuracy and low latency for accurate and fast responsiveness. Many efforts have been made to develop end-to-end neural networks, in which depthwise separable convolutions, temporal convolutions, and LSTMs are adopted as building units. Nonetheless, these networks designed with human expertise may not achieve an optimal trade-off in an expansive search space. In this paper, we propose to leverage recent advances in differentiable neural architecture search to discover more efficient networks. Our searched model attains 97.2% top-1 accuracy on Google Speech Command Dataset v1 with only nearly 100K parameters., Comment: ICASSP 2021

Attention-based seq2seq text-to-speech systems, especially those use self-attention networks (SAN), have achieved state-of-art performance. But an expressive corpus with rich prosody is still challenging to model as 1) prosodic aspects, which span across different sentential granularities and mainly determine acoustic expressiveness, are difficult to quantize and label and 2) the current seq2seq framework extracts prosodic information solely from a text encoder, which is easily collapsed to an averaged expression for expressive contents. In this paper, we propose a context extractor, which is built upon SAN-based text encoder, to sufficiently exploit the sentential context over an expressive corpus for seq2seq-based TTS. Our context extractor first collects prosodic-related sentential context information from different SAN layers and then aggregates them to learn a comprehensive sentence representation to enhance the expressiveness of the final generated speech. Specifically, we investigate two methods of context aggregation: 1) direct aggregation which directly concatenates the outputs of different SAN layers, and 2) weighted aggregation which uses multi-head attention to automatically learn contributions for different SAN layers. Experiments on two expressive corpora show that our approach can produce more natural speech with much richer prosodic variations, and weighted aggregation is more superior in modeling expressivity., Comment: Accepted by Interspeech2020

Neural network-based vocoders have recently demonstrated the powerful ability to synthesize high-quality speech. These models usually generate samples by conditioning on spectral features, such as Mel-spectrogram and fundamental frequency, which is crucial to speech synthesis. However, the feature extraction procession tends to depend heavily on human knowledge resulting in a less expressive description of the origin audio. In this work, we proposed RawNet, a complete end-to-end neural vocoder following the auto-encoder structure for speaker-dependent and -independent speech synthesis. It automatically learns to extract features and recover audio using neural networks, which include a coder network to capture a higher representation of the input audio and an autoregressive voder network to restore the audio in a sample-by-sample manner. The coder and voder are jointly trained directly on the raw waveform without any human-designed features. The experimental results show that RawNet achieves a better speech quality using a simplified model architecture and obtains a faster speech generation speed at the inference stage.

In this paper, we propose an attention-based end-to-end model for multi-channel keyword spotting (KWS), which is trained to optimize the KWS result directly. As a result, our model outperforms the baseline model with signal pre-processing techniques in both the clean and noisy testing data. We also found that multi-task learning results in a better performance when the training and testing data are similar. Transfer learning and multi-target spectral mapping can dramatically enhance the robustness to the noisy environment. At 0.1 false alarm (FA) per hour, the model with transfer learning and multi-target mapping gain an absolute 30% improvement in the wake-up rate in the noisy data with SNR about -20., Comment: Submitted to ICASSP 2019

In this paper, we propose a sequence-to-sequence model for keyword spotting (KWS). Compared with other end-to-end architectures for KWS, our model simplifies the pipelines of production-quality KWS system and satisfies the requirement of high accuracy, low-latency, and small-footprint. We also evaluate the performances of different encoder architectures, which include LSTM and GRU. Experiments on the real-world wake-up data show that our approach outperforms the recently proposed attention-based end-to-end model. Specifically speaking, with 73K parameters, our sequence-to-sequence model achieves $\sim$3.05\% false rejection rate (FRR) at 0.1 false alarm (FA) per hour., Comment: Submitted to ICASSP 2019

In this paper, we propose an attention-based end-to-end neural approach for small-footprint keyword spotting (KWS), which aims to simplify the pipelines of building a production-quality KWS system. Our model consists of an encoder and an attention mechanism. The encoder transforms the input signal into a high level representation using RNNs. Then the attention mechanism weights the encoder features and generates a fixed-length vector. Finally, by linear transformation and softmax function, the vector becomes a score used for keyword detection. We also evaluate the performance of different encoder architectures, including LSTM, GRU and CRNN. Experiments on real-world wake-up data show that our approach outperforms the recent Deep KWS approach by a large margin and the best performance is achieved by CRNN. To be more specific, with ~84K parameters, our attention-based model achieves 1.02% false rejection rate (FRR) at 1.0 false alarm (FA) per hour., Comment: attention-based model, end-to-end keyword spotting, convolutional neural networks, recurrent neural networks

Speaker adaptation aims to estimate a speaker specific acoustic model from a speaker independent one to minimize the mismatch between the training and testing conditions arisen from speaker variabilities. A variety of neural network adaptation methods have been proposed since deep learning models have become the main stream. But there still lacks an experimental comparison between different methods, especially when DNN-based acoustic models have been advanced greatly. In this paper, we aim to close this gap by providing an empirical evaluation of three typical speaker adaptation methods: LIN, LHUC and KLD. Adaptation experiments, with different size of adaptation data, are conducted on a strong TDNN-LSTM acoustic model. More challengingly, here, the source and target we are concerned with are standard Mandarin speaker model and accented Mandarin speaker model. We compare the performances of different methods and their combinations. Speaker adaptation performance is also examined by speaker's accent degree., Comment: Interspeech 2018

We investigate the use of generative adversarial networks (GANs) in speech dereverberation for robust speech recognition. GANs have been recently studied for speech enhancement to remove additive noises, but there still lacks of a work to examine their ability in speech dereverberation and the advantages of using GANs have not been fully established. In this paper, we provide deep investigations in the use of GAN-based dereverberation front-end in ASR. First, we study the effectiveness of different dereverberation networks (the generator in GAN) and find that LSTM leads a significant improvement as compared with feed-forward DNN and CNN in our dataset. Second, further adding residual connections in the deep LSTMs can boost the performance as well. Finally, we find that, for the success of GAN, it is important to update the generator and the discriminator using the same mini-batch data during training. Moreover, using reverberant spectrogram as a condition to discriminator, as suggested in previous studies, may degrade the performance. In summary, our GAN-based dereverberation front-end achieves 14%-19% relative CER reduction as compared to the baseline DNN dereverberation network when tested on a strong multi-condition training acoustic model., Comment: Interspeech 2018