As the cornerstone of other important technologies, such as speech recognition and speech synthesis, speech enhancement is a critical area in audio signal processing. In this paper, a new deep learning structure for speech enhancement is demonstrated. The model introduces a "full" attention mechanism to a bidirectional sequence-to-sequence method to make use of latent information after each focal frame. This is an extension of the previous attention-based RNN method. The proposed bidirectional attention-based architecture achieves better performance in terms of speech quality (PESQ), compared with OM-LSA, CNN-LSTM, T-GSA and the unidirectional attention-based LSTM baseline.
Teleconferencing is becoming essential during the COVID-19 pandemic. However, in real-world applications, speech quality can deteriorate due to, for example, background interference, noise, or reverberation. To solve this problem, target speech extraction from the mixture signals can be performed with the aid of the user's vocal features. Various features are accounted for in this study's proposed system, including speaker embeddings derived from user enrollment and a novel long-short-term spatial coherence (LSTSC) feature to the target speaker activity. As a learning-based approach, a target speech sifting network was employed to extract the relevant features. The network trained with LSTSC in the proposed approach is robust to microphone array geometries and the number of microphones. Furthermore, the proposed enhancement system was compared with a baseline system with speaker embeddings and interchannel phase difference. The results demonstrated the superior performance of the proposed system over the baseline in enhancement performance and robustness.
Speech Translation (ST) is the task of translating speech in one language into text in another language. Traditional cascaded approaches for ST, using Automatic Speech Recognition (ASR) and Machine Translation (MT) systems, are prone to error propagation. End-to-end approaches use only one system to avoid propagating error, yet are difficult to employ due to data scarcity. We explore zero-shot translation, which enables translating a pair of languages that is unseen during training, thus avoid the use of end-to-end ST data. Zero-shot translation has been shown to work for multilingual machine translation, yet has not been studied for speech translation. We attempt to build zero-shot ST models that are trained only on ASR and MT tasks but can do ST task during inference. The challenge is that the representation of text and audio is significantly different, thus the models learn ASR and MT tasks in different ways, making it non-trivial to perform zero-shot. These models tend to output the wrong language when performing zero-shot ST. We tackle the issues by including additional training data and an auxiliary loss function that minimizes the text-audio difference. Our experiment results and analysis show that the methods are promising for zero-shot ST. Moreover, our methods are particularly useful in the few-shot settings where a limited amount of ST data is available, with improvements of up to +11.8 BLEU points compared to direct end-to-end ST models and +3.9 BLEU points compared to ST models fine-tuned from pre-trained ASR model.
Self-supervised learning (SSL) methods have proven to be very successful in automatic speech recognition (ASR). These great improvements have been reported mostly based on highly curated datasets such as LibriSpeech for non-streaming End-to-End ASR models. However, the pivotal characteristics of SSL is to be utilized for any untranscribed audio data. In this paper, we provide a full exploration on how to utilize uncurated audio data in SSL from data pre-processing to deploying an streaming hybrid ASR model. More specifically, we present (1) the effect of Audio Event Detection (AED) model in data pre-processing pipeline (2) analysis on choosing optimizer and learning rate scheduling (3) comparison of recently developed contrastive losses, (4) comparison of various pre-training strategies such as utilization of in-domain versus out-domain pre-training data, monolingual versus multilingual pre-training data, multi-head multilingual SSL versus single-head multilingual SSL and supervised pre-training versus SSL. The experimental results show that SSL pre-training with in-domain uncurated data can achieve better performance in comparison to all the alternative out-domain pre-training strategies.
In this work, we propose a speaker anonymization pipeline that leverages high quality automatic speech recognition and synthesis systems to generate speech conditioned on phonetic transcriptions and anonymized speaker embeddings. Using phones as the intermediate representation ensures near complete elimination of speaker identity information from the input while preserving the original phonetic content as much as possible. Our experimental results on LibriSpeech and VCTK corpora reveal two key findings: 1) although automatic speech recognition produces imperfect transcriptions, our neural speech synthesis system can handle such errors, making our system feasible and robust, and 2) combining speaker embeddings from different resources is beneficial and their appropriate normalization is crucial. Overall, our final best system outperforms significantly the baselines provided in the Voice Privacy Challenge 2020 in terms of privacy robustness against a lazy-informed attacker while maintaining high intelligibility and naturalness of the anonymized speech.
We aim to solve the highly challenging task of generating continuous sign language videos solely from speech segments for the first time. Recent efforts in this space have focused on generating such videos from human-annotated text transcripts without considering other modalities. However, replacing speech with sign language proves to be a practical solution while communicating with people suffering from hearing loss. Therefore, we eliminate the need of using text as input and design techniques that work for more natural, continuous, freely uttered speech covering an extensive vocabulary. Since the current datasets are inadequate for generating sign language directly from speech, we collect and release the first Indian sign language dataset comprising speech-level annotations, text transcripts, and the corresponding sign-language videos. Next, we propose a multi-tasking transformer network trained to generate signer's poses from speech segments. With speech-to-text as an auxiliary task and an additional cross-modal discriminator, our model learns to generate continuous sign pose sequences in an end-to-end manner. Extensive experiments and comparisons with other baselines demonstrate the effectiveness of our approach. We also conduct additional ablation studies to analyze the effect of different modules of our network. A demo video containing several results is attached to the supplementary material.
In speaker recognition, where speech segments are mapped to embeddings on the unit hypersphere, two scoring back-ends are commonly used, namely cosine scoring and PLDA. We have recently proposed PSDA, an analog to PLDA that uses Von Mises-Fisher distributions instead of Gaussians. In this paper, we present toroidal PSDA (T-PSDA). It extends PSDA with the ability to model within and between-speaker variabilities in toroidal submanifolds of the hypersphere. Like PLDA and PSDA, the model allows closed-form scoring and closed-form EM updates for training. On VoxCeleb, we find T-PSDA accuracy on par with cosine scoring, while PLDA accuracy is inferior. On NIST SRE'21 we find that T-PSDA gives large accuracy gains compared to both cosine scoring and PLDA.
Frequency domain processing, and in particular the use of Modified Discrete Cosine Transform (MDCT), is the most widespread approach to audio coding. However, at low bitrates, audio quality, especially for speech, degrades drastically due to the lack of available bits to directly code the transform coefficients. Traditionally, post-filtering has been used to mitigate artefacts in the coded speech by exploiting a-priori information of the source and extra transmitted parameters. Recently, data-driven post-filters have shown better results, but at the cost of significant additional complexity and delay. In this work, we propose a mask-based post-filter operating directly in MDCT domain of the codec, inducing no extra delay. The real-valued mask is applied to the quantized MDCT coefficients and is estimated from a relatively lightweight convolutional encoder-decoder network. Our solution is tested on the recently standardized low-delay, low-complexity codec (LC3) at lowest possible bitrate of 16 kbps. Objective and subjective assessments clearly show the advantage of this approach over the conventional post-filter, with an average improvement of 10 MUSHRA points over the LC3 coded speech.
This study addresses robust automatic speech recognition (ASR) by introducing a Conformer-based acoustic model. The proposed model builds on a state-of-the-art recognition system using a bi-directional long short-term memory (BLSTM) model with utterance-wise dropout and iterative speaker adaptation, but employs a Conformer encoder instead of the BLSTM network. The Conformer encoder uses a convolution-augmented attention mechanism for acoustic modeling. The proposed system is evaluated on the monaural ASR task of the CHiME-4 corpus. Coupled with utterance-wise normalization and speaker adaptation, our model achieves $6.25\%$ word error rate, which outperforms the previous best system by $8.4\%$ relatively. In addition, the proposed Conformer-based model is $18.3\%$ smaller in model size and reduces total training time by $79.6\%$.
This paper proposes a novel approach for speech signal prediction based on a recurrent neural network (RNN). Unlike existing RNN-based predictors, which operate on parametric features and are trained offline on a large collection of such features, the proposed predictor operates directly on speech samples and is trained online on the recent past of the speech signal. Optionally, the network can be pre-trained offline to speed-up convergence at start-up. The proposed predictor is a single end-to-end network that captures all sorts of dependencies between samples, and therefore has the potential to outperform classical linear/non-linear and short-term/long-term speech predictor structures. We apply it to the packet loss concealment (PLC) problem and show that it outperforms the standard ITU G.711 Appendix I PLC technique.