An independent, automated method of decoding and transcribing oral speech is known as automatic speech recognition (ASR). A typical ASR system extracts feature from audio recordings or streams and run one or more algorithms to map the features to corresponding texts. Numerous of research has been done in the field of speech signal processing in recent years. When given adequate resources, both conventional ASR and emerging end-to-end (E2E) speech recognition have produced promising results. However, for low-resource languages like Bengali, the current state of ASR lags behind, although the low resource state does not reflect upon the fact that this language is spoken by over 500 million people all over the world. Despite its popularity, there aren't many diverse open-source datasets available, which makes it difficult to conduct research on Bengali speech recognition systems. This paper is a part of the competition named `BUET CSE Fest DL Sprint'. The purpose of this paper is to improve the speech recognition performance of the Bengali language by adopting speech recognition technology on the E2E structure based on the transfer learning framework. The proposed method effectively models the Bengali language and achieves 3.819 score in `Levenshtein Mean Distance' on the test dataset of 7747 samples, when only 1000 samples of train dataset were used to train.
Generating talking person portraits with arbitrary speech audio is a crucial problem in the field of digital human and metaverse. A modern talking face generation method is expected to achieve the goals of generalized audio-lip synchronization, good video quality, and high system efficiency. Recently, neural radiance field (NeRF) has become a popular rendering technique in this field since it could achieve high-fidelity and 3D-consistent talking face generation with a few-minute-long training video. However, there still exist several challenges for NeRF-based methods: 1) as for the lip synchronization, it is hard to generate a long facial motion sequence of high temporal consistency and audio-lip accuracy; 2) as for the video quality, due to the limited data used to train the renderer, it is vulnerable to out-of-domain input condition and produce bad rendering results occasionally; 3) as for the system efficiency, the slow training and inference speed of the vanilla NeRF severely obstruct its usage in real-world applications. In this paper, we propose GeneFace++ to handle these challenges by 1) utilizing the pitch contour as an auxiliary feature and introducing a temporal loss in the facial motion prediction process; 2) proposing a landmark locally linear embedding method to regulate the outliers in the predicted motion sequence to avoid robustness issues; 3) designing a computationally efficient NeRF-based motion-to-video renderer to achieves fast training and real-time inference. With these settings, GeneFace++ becomes the first NeRF-based method that achieves stable and real-time talking face generation with generalized audio-lip synchronization. Extensive experiments show that our method outperforms state-of-the-art baselines in terms of subjective and objective evaluation. Video samples are available at https://genefaceplusplus.github.io .
Most state-of-the-art Text-to-Speech systems use the mel-spectrogram as an intermediate representation, to decompose the task into acoustic modelling and waveform generation. A mel-spectrogram is extracted from the waveform by a simple, fast DSP operation, but generating a high-quality waveform from a mel-spectrogram requires computationally expensive machine learning: a neural vocoder. Our proposed ``autovocoder'' reverses this arrangement. We use machine learning to obtain a representation that replaces the mel-spectrogram, and that can be inverted back to a waveform using simple, fast operations including a differentiable implementation of the inverse STFT. The autovocoder generates a waveform 5 times faster than the DSP-based Griffin-Lim algorithm, and 14 times faster than the neural vocoder HiFi-GAN. We provide perceptual listening test results to confirm that the speech is of comparable quality to HiFi-GAN in the copy synthesis task.
The Transformer architecture model, based on self-attention and multi-head attention, has achieved remarkable success in offline end-to-end Automatic Speech Recognition (ASR). However, self-attention and multi-head attention cannot be easily applied for streaming or online ASR. For self-attention in Transformer ASR, the softmax normalization function-based attention mechanism makes it impossible to highlight important speech information. For multi-head attention in Transformer ASR, it is not easy to model monotonic alignments in different heads. To overcome these two limits, we integrate sparse attention and monotonic attention into Transformer-based ASR. The sparse mechanism introduces a learned sparsity scheme to enable each self-attention structure to fit the corresponding head better. The monotonic attention deploys regularization to prune redundant heads for the multi-head attention structure. The experiments show that our method can effectively improve the attention mechanism on widely used benchmarks of speech recognition.
Recently, diffusion models (DMs) have been increasingly used in audio processing tasks, including speech super-resolution (SR), which aims to restore high-frequency content given low-resolution speech utterances. This is commonly achieved by conditioning the network of noise predictor with low-resolution audio. In this paper, we propose a novel sampling algorithm that communicates the information of the low-resolution audio via the reverse sampling process of DMs. The proposed method can be a drop-in replacement for the vanilla sampling process and can significantly improve the performance of the existing works. Moreover, by coupling the proposed sampling method with an unconditional DM, i.e., a DM with no auxiliary inputs to its noise predictor, we can generalize it to a wide range of SR setups. We also attain state-of-the-art results on the VCTK Multi-Speaker benchmark with this novel formulation.
Despite the success of deep neural network (DNN) on sequential data (i.e., scene text and speech) recognition, it suffers from the over-confidence problem mainly due to overfitting in training with the cross-entropy loss, which may make the decision-making less reliable. Confidence calibration has been recently proposed as one effective solution to this problem. Nevertheless, the majority of existing confidence calibration methods aims at non-sequential data, which is limited if directly applied to sequential data since the intrinsic contextual dependency in sequences or the class-specific statistical prior is seldom exploited. To the end, we propose a Context-Aware Selective Label Smoothing (CASLS) method for calibrating sequential data. The proposed CASLS fully leverages the contextual dependency in sequences to construct confusion matrices of contextual prediction statistics over different classes. Class-specific error rates are then used to adjust the weights of smoothing strength in order to achieve adaptive calibration. Experimental results on sequence recognition tasks, including scene text recognition and speech recognition, demonstrate that our method can achieve the state-of-the-art performance.
A two-stage approach is proposed for speaker counting and speech separation in noisy and reverberant environments. A spatial coherence matrix (SCM) is computed using whitened relative transfer functions (wRTFs) across time frames. The global activity functions of each speaker are estimated on the basis of a simplex constructed using the eigenvectors of the SCM, while the local coherence functions are computed from the coherence between the wRTFs of a time-frequency bin and the global activity function-weighted RTF of the target speaker. In speaker counting, we use the eigenvalues of the SCM and the maximum similarity of the interframe global activity distributions between two speakers as the input features to the speaker counting network (SCnet). In speaker separation, a global and local activity-driven network (GLADnet) is utilized to estimate a speaker mask, which is particularly useful for highly overlapping speech signals. Experimental results obtained from the real meeting recordings demonstrated the superior speaker counting and speaker separation performance achieved by the proposed learning-based system without prior knowledge of the array configurations.
When recognizing emotions from speech, we encounter two common problems: how to optimally capture emotion-relevant information from the speech signal and how to best quantify or categorize the noisy subjective emotion labels. Self-supervised pre-trained representations can robustly capture information from speech enabling state-of-the-art results in many downstream tasks including emotion recognition. However, better ways of aggregating the information across time need to be considered as the relevant emotion information is likely to appear piecewise and not uniformly across the signal. For the labels, we need to take into account that there is a substantial degree of noise that comes from the subjective human annotations. In this paper, we propose a novel approach to attentive pooling based on correlations between the representations' coefficients combined with label smoothing, a method aiming to reduce the confidence of the classifier on the training labels. We evaluate our proposed approach on the benchmark dataset IEMOCAP, and demonstrate high performance surpassing that in the literature. The code to reproduce the results is available at github.com/skakouros/s3prl_attentive_correlation.
A primary objective of news articles is to establish the factual record for an event, frequently achieved by conveying both the details of the specified event (i.e., the 5 Ws; Who, What, Where, When and Why regarding the event) and how people reacted to it (i.e., reported statements). However, existing work on news summarization almost exclusively focuses on the event details. In this work, we propose the novel task of summarizing the reactions of different speakers, as expressed by their reported statements, to a given event. To this end, we create a new multi-document summarization benchmark, SUMREN, comprising 745 summaries of reported statements from various public figures obtained from 633 news articles discussing 132 events. We propose an automatic silver training data generation approach for our task, which helps smaller models like BART achieve GPT-3 level performance on this task. Finally, we introduce a pipeline-based framework for summarizing reported speech, which we empirically show to generate summaries that are more abstractive and factual than baseline query-focused summarization approaches.
We present our latest findings on backchannel modeling novelly motivated by the canonical use of the minimal responses Yeah and Uh-huh in English and their correspondent tokens in German, and the effect of encoding the speaker-listener interaction. Backchanneling theories emphasize the active and continuous role of the listener in the course of the conversation, their effects on the speaker's subsequent talk, and the consequent dynamic speaker-listener interaction. Therefore, we propose a neural-based acoustic backchannel classifier on minimal responses by processing acoustic features from the speaker speech, capturing and imitating listeners' backchanneling behavior, and encoding speaker-listener interaction. Our experimental results on the Switchboard and GECO datasets reveal that in almost all tested scenarios the speaker or listener behavior embeddings help the model make more accurate backchannel predictions. More importantly, a proper interaction encoding strategy, i.e., combining the speaker and listener embeddings, leads to the best performance on both datasets in terms of F1-score.