In this paper, we investigate the manner in which interpretable sub-word speech units emerge within a convolutional neural network model trained to associate raw speech waveforms with semantically related natural image scenes. We show how diphone boundaries can be superficially extracted from the activation patterns of intermediate layers of the model, suggesting that the model may be leveraging these events for the purpose of word recognition. We present a series of experiments investigating the information encoded by these events.
Ensuring proper punctuation and letter casing is a key pre-processing step towards applying complex natural language processing algorithms. This is especially significant for textual sources where punctuation and casing are missing, such as the raw output of automatic speech recognition systems. Additionally, short text messages and micro-blogging platforms offer unreliable and often wrong punctuation and casing. This survey offers an overview of both historical and state-of-the-art techniques for restoring punctuation and correcting word casing. Furthermore, current challenges and research directions are highlighted.
In this paper, we propose a high-quality generative text-to-speech (TTS) system using an effective spectrum and excitation estimation method. Our previous research verified the effectiveness of the ExcitNet-based speech generation model in a parametric TTS framework. However, the challenge remains to build a high-quality speech synthesis system because auxiliary conditional features estimated by a simple deep neural network often contain large prediction errors, and the errors are inevitably propagated throughout the autoregressive generation process of the ExcitNet vocoder. To generate more natural speech signals, we exploited a sequence-to-sequence (seq2seq) acoustic model with an attention-based generative network (e.g., Tacotron 2) to estimate the condition parameters of the ExcitNet vocoder. Because the seq2seq acoustic model accurately estimates spectral parameters, and because the ExcitNet model effectively generates the corresponding time-domain excitation signals, combining these two models can synthesize natural speech signals. Furthermore, we verified the merit of the proposed method in producing expressive speech segments by adopting a global style token-based emotion embedding method. The experimental results confirmed that the proposed system significantly outperforms the systems with a similarly configured conventional WaveNet vocoder and our best prior parametric TTS counterpart.
Medication errors most commonly occur at the ordering or prescribing stage, potentially leading to medical complications and poor health outcomes. While it is possible to catch these errors using different techniques; the focus of this work is on textual and contextual analysis of prescription information to detect and prevent potential medication errors. In this paper, we demonstrate how to use BERT-based contextual language models to detect anomalies in written or spoken text based on a data set extracted from real-world medical data of thousands of patient records. The proposed models are able to learn patterns of text dependency and predict erroneous output based on contextual information such as patient data. The experimental results yield accuracy up to 96.63% for text input and up to 79.55% for speech input, which is satisfactory for most real-world applications.
In this paper, we propose a single-channel speech dereverberation system (DeReGAT) based on convolutional, bidirectional long short-term memory and deep feed-forward neural network (CBLDNN) with generative adversarial training (GAT). In order to obtain better speech quality instead of only minimizing a mean square error (MSE), GAT is employed to make the dereverberated speech indistinguishable form the clean samples. Besides, our system can deal with wide range reverberation and be well adapted to variant environments. The experimental results show that the proposed model outperforms weighted prediction error (WPE) and deep neural network-based systems. In addition, DeReGAT is extended to an online speech dereverberation scenario, which reports comparable performance with the offline case.
We investigate the automatic processing of child speech therapy sessions using ultrasound visual biofeedback, with a specific focus on complementing acoustic features with ultrasound images of the tongue for the tasks of speaker diarization and time-alignment of target words. For speaker diarization, we propose an ultrasound-based time-domain signal which we call estimated tongue activity. For word-alignment, we augment an acoustic model with low-dimensional representations of ultrasound images of the tongue, learned by a convolutional neural network. We conduct our experiments using the Ultrasuite repository of ultrasound and speech recordings for child speech therapy sessions. For both tasks, we observe that systems augmented with ultrasound data outperform corresponding systems using only the audio signal.
Speech translation has traditionally been approached through cascaded models consisting of a speech recognizer trained on a corpus of transcribed speech, and a machine translation system trained on parallel texts. Several recent works have shown the feasibility of collapsing the cascade into a single, direct model that can be trained in an end-to-end fashion on a corpus of translated speech. However, experiments are inconclusive on whether the cascade or the direct model is stronger, and have only been conducted under the unrealistic assumption that both are trained on equal amounts of data, ignoring other available speech recognition and machine translation corpora. In this paper, we demonstrate that direct speech translation models require more data to perform well than cascaded models, and while they allow including auxiliary data through multi-task training, they are poor at exploiting such data, putting them at a severe disadvantage. As a remedy, we propose the use of end-to-end trainable models with two attention mechanisms, the first establishing source speech to source text alignments, the second modeling source to target text alignment. We show that such models naturally decompose into multi-task-trainable recognition and translation tasks and propose an attention-passing technique that alleviates error propagation issues in a previous formulation of a model with two attention stages. Our proposed model outperforms all examined baselines and is able to exploit auxiliary training data much more effectively than direct attentional models.
Recent advances in text-to-speech (TTS) led to the development of flexible multi-speaker end-to-end TTS systems. We extend state-of-the-art attention-based automatic speech recognition (ASR) systems with synthetic audio generated by a TTS system trained only on the ASR corpora itself. ASR and TTS systems are built separately to show that text-only data can be used to enhance existing end-to-end ASR systems without the necessity of parameter or architecture changes. We compare our method with language model integration of the same text data and with simple data augmentation methods like SpecAugment and show that performance improvements are mostly independent. We achieve improvements of up to 33% relative in word-error-rate (WER) over a strong baseline with data-augmentation in a low-resource environment (LibriSpeech-100h), closing the gap to a comparable oracle experiment by more than 50\%. We also show improvements of up to 5% relative WER over our most recent ASR baseline on LibriSpeech-960h.
Recent video and language pretraining frameworks lack the ability to generate sentences. We present Multimodal Video Generative Pretraining (MV-GPT), a new pretraining framework for learning from unlabelled videos which can be effectively used for generative tasks such as multimodal video captioning. Unlike recent video-language pretraining frameworks, our framework trains both a multimodal video encoder and a sentence decoder jointly. To overcome the lack of captions in unlabelled videos, we leverage the future utterance as an additional text source and propose a bidirectional generation objective -- we generate future utterances given the present mulitmodal context, and also the present utterance given future observations. With this objective, we train an encoder-decoder model end-to-end to generate a caption from raw pixels and transcribed speech directly. Our model achieves state-of-the-art performance for multimodal video captioning on four standard benchmarks, as well as for other video understanding tasks such as VideoQA, video retrieval and action classification.
Recent advances in deep learning show that end-to-end speech to text translation model is a promising approach to direct the speech translation field. In this work, we provide an overview of different end-to-end architectures, as well as the usage of an auxiliary connectionist temporal classification (CTC) loss for better convergence. We also investigate on pre-training variants such as initializing different components of a model using pre-trained models, and their impact on the final performance, which gives boosts up to 4% in BLEU and 5% in TER. Our experiments are performed on 270h IWSLT TED-talks En->De, and 100h LibriSpeech Audiobooks En->Fr. We also show improvements over the current end-to-end state-of-the-art systems on both tasks.