This paper presents a novel data augmentation technique for text-to-speech (TTS), that allows to generate new (text, audio) training examples without requiring any additional data. Our goal is to in-crease diversity of text conditionings available during training. This helps to reduce overfitting, especially in low-resource settings. Our method relies on substituting text and audio fragments in a way that preserves syntactical correctness. We take additional measures to ensure that synthesized speech does not contain artifacts caused by combining inconsistent audio samples. The perceptual evaluations show that our method improves speech quality over a number of datasets, speakers, and TTS architectures. We also demonstrate that it greatly improves robustness of attention-based TTS models.

The state-of-the-art in text-to-speech (TTS) synthesis has recently improved considerably due to novel neural waveform generation methods, such as WaveNet. However, these methods suffer from their slow sequential inference process, while their parallel versions are difficult to train and even more computationally expensive. Meanwhile, generative adversarial networks (GANs) have achieved impressive results in image generation and are making their way into audio applications; parallel inference is among their lucrative properties. By adopting recent advances in GAN training techniques, this investigation studies waveform generation for TTS in two domains (speech signal and glottal excitation). Listening test results show that while direct waveform generation with GAN is still far behind WaveNet, a GAN-based glottal excitation model can achieve quality and voice similarity on par with a WaveNet vocoder.

Currently, there is increasing interest to use sequence-to-sequence models in text-to-speech (TTS) synthesis with attention like that in Tacotron models. These models are end-to-end, meaning that they learn both co-articulation and duration properties directly from text and speech. Since these models are entirely data-driven, they need large amounts of data to generate synthetic speech of good quality. However, in challenging speaking styles, such as Lombard speech, it is difficult to record sufficiently large speech corpora. Therefore, we propose a transfer learning method to adapt a TTS system of normal speaking style to Lombard style. We also experiment with a WaveNet vocoder along with a traditional vocoder (WORLD) in the synthesis of Lombard speech. The subjective and objective evaluation results indicated that the proposed adaptation system coupled with the WaveNet vocoder clearly outperformed the conventional deep neural network based TTS system in the synthesis of Lombard speech

Recent advances in neural network -based text-to-speech have reached human level naturalness in synthetic speech. The present sequence-to-sequence models can directly map text to mel-spectrogram acoustic features, which are convenient for modeling, but present additional challenges for vocoding (i.e., waveform generation from the acoustic features). Highquality synthesis can be achieved with neural vocoders, such as WaveNet, but such autoregressive models suffer from slow sequential inference. Meanwhile, their existing parallel inference counterparts are difficult to train and require increasingly large model sizes. In this paper, we propose an alternative training strategy for a parallel neural vocoder utilizing generative adversarial networks, and integrate a linear predictive synthesis filter into the model. Results show that the proposed model achieves significant improvement in inference speed, while outperforming a WaveNet in copy-synthesis quality.

Recently, generative neural network models which operate directly on raw audio, such as WaveNet, have improved the state of the art in text-to-speech synthesis (TTS). Moreover, there is increasing interest in using these models as statistical vocoders for generating speech waveforms from various acoustic features. However, there is also a need to reduce the model complexity, without compromising the synthesis quality. Previously, glottal pulseforms (i.e., time-domain waveforms corresponding to the source of human voice production mechanism) have been successfully synthesized in TTS by glottal vocoders using straightforward deep feedforward neural networks. Therefore, it is natural to extend the glottal waveform modeling domain to use the more powerful WaveNet-like architecture. Furthermore, due to their inherent simplicity, glottal excitation waveforms permit scaling down the waveform generator architecture. In this study, we present a raw waveform glottal excitation model, called GlotNet, and compare its performance with the corresponding direct speech waveform model, WaveNet, using equivalent architectures. The models are evaluated as part of a statistical parametric TTS system. Listening test results show that both approaches are rated highly in voice similarity to the target speaker, and obtain similar quality ratings with large models. Furthermore, when the model size is reduced, the quality degradation is less severe for GlotNet.

This paper proposes a method for generating speech from filterbank mel frequency cepstral coefficients (MFCC), which are widely used in speech applications, such as ASR, but are generally considered unusable for speech synthesis. First, we predict fundamental frequency and voicing information from MFCCs with an autoregressive recurrent neural net. Second, the spectral envelope information contained in MFCCs is converted to all-pole filters, and a pitch-synchronous excitation model matched to these filters is trained. Finally, we introduce a generative adversarial network -based noise model to add a realistic high-frequency stochastic component to the modeled excitation signal. The results show that high quality speech reconstruction can be obtained, given only MFCC information at test time.

Recent speech technology research has seen a growing interest in using WaveNets as statistical vocoders, i.e., generating speech waveforms from acoustic features. These models have been shown to improve the generated speech quality over classical vocoders in many tasks, such as text-to-speech synthesis and voice conversion. Furthermore, conditioning WaveNets with acoustic features allows sharing the waveform generator model across multiple speakers without additional speaker codes. However, multi-speaker WaveNet models require large amounts of training data and computation to cover the entire acoustic space. This paper proposes leveraging the source-filter model of speech production to more effectively train a speaker-independent waveform generator with limited resources. We present a multi-speaker 'GlotNet' vocoder, which utilizes a WaveNet to generate glottal excitation waveforms, which are then used to excite the corresponding vocal tract filter to produce speech. Listening tests show that the proposed model performs favourably to a direct WaveNet vocoder trained with the same model architecture and data.

A vocoder is used to express a speech waveform with a controllable parametric representation that can be converted back into a speech waveform. Vocoders representing their main categories (mixed excitation, glottal, and sinusoidal vocoders) were compared in this study with formal and crowd-sourced listening tests. The vocoder quality was measured within the context of analysis-synthesis as well as text-to-speech (TTS) synthesis in a modern statistical parametric speech synthesis framework. Furthermore, the TTS experiments were divided into synthesis with vocoder-specific features and synthesis with a shared envelope model, where the waveform generation method of the vocoders is mainly responsible for the quality differences. Finally, all of the tests included four distinct voices as a way to investigate the effect of different speakers on the synthesized speech quality. The obtained results suggest that the choice of the voice has a profound impact on the overall quality of the vocoder-generated speech, and the best vocoder for each voice can vary case by case. The single best-rated TTS system was obtained with the glottal vocoder GlottDNN using a male voice with low expressiveness. However, the results indicate that the sinusoidal vocoder PML (pulse model in log-domain) has the best overall performance across the performed tests. Finally, when controlling for the spectral models of the vocoders, the observed differences are similar to the baseline results. This indicates that the waveform generation method of a vocoder is essential for quality improvements.

Auto-regressive modeling is a prevalent source-filter separation method of speech. Conventional linear prediction (LP) and its derivatives such as weighted linear prediction (WeLP) produce parametric spectral models within a linear frequency scale, whereas frequency-warped linear prediction (WaLP) can be used to take into account the frequency sensitivity of the human auditory system. From the perspective of glottal vocoding, the principles behind WeLP have been found to be beneficial for an accurate separation of the glottal source signal and the vocal tract transfer function, but this approach can not utilize the auditory benefits of frequency warping. On the other hand, the WaLP approach suffers from less accurate source-filter separation properties. In this study, a generalized frequency-warped time-weighted linear prediction (WWLP) analysis is proposed. Experiments with WWLP are performed within the context of glottal vocoding. The subjective listening test results show that WWLP-based spectral envelope modeling is able to increase quality over previously developed methods in some of the test cases.

In statistical parametric speech synthesis (SPSS), a few studies have investigated the Lombard effect, specifically by using hidden Markov model (HMM)-based systems. Recently, artificial neural networks have demonstrated promising results in SPSS, specifically by using long short-term memory recurrent neural networks (LSTMs). The Lombard effect, however, has not been studied in the LSTM-based speech synthesis systems. In this study, we propose three methods for Lombard speech adaptation in LSTM-based speech synthesis. In particular, (1) we augment Lombard specific information with the linguistic features as input, (2) scale the hidden activations using the learning hidden unit contributions (LHUC) method, and (3) fine-tune the LSTMs trained on normal speech with a small Lombard speech data. To investigate the effectiveness of the proposed methods, we carry out experiments using small (10 utterances) and large (500 utterances) Lombard speech data. Experimental results confirm the adaptability of the LSTMs, and similarity tests show that the LSTMs can achieve significantly better adaptation performance than the HMMs in both small and large data conditions.

Recent studies have shown that text-to-speech synthesis quality can be improved by using glottal vocoding. This refers to vocoders that parameterize speech into two parts, the glottal excitation and vocal tract, that occur in the human speech production apparatus. Current glottal vocoders generate the glottal excitation waveform by using deep neural networks (DNNs). However, the squared error-based training of the present glottal excitation models is limited to generating conditional average waveforms, which fails to capture the stochastic variation of the waveforms. As a result, shaped noise is added as post-processing. In this study, we propose a new method for predicting glottal waveforms by generative adversarial networks (GANs). GANs are generative models that aim to embed the data distribution in a latent space, enabling generation of new instances very similar to the original by randomly sampling the latent distribution. The glottal pulses generated by GANs show a stochastic component similar to natural glottal pulses. In our experiments, we compare synthetic speech generated using glottal waveforms produced by both DNNs and GANs. The results show that the newly proposed GANs achieve synthesis quality comparable to that of widely-used DNNs, without using an additive noise component.

Neural network-based models that generate glottal excitation waveforms from acoustic features have been found to give improved quality in statistical parametric speech synthesis. Until now, however, these models have been trained separately from the acoustic model. This creates mismatch between training and synthesis, as the synthesized acoustic features used for the excitation model input differ from the original inputs, with which the model was trained on. Furthermore, due to the errors in predicting the vocal tract filter, the original excitation waveforms do not provide perfect reconstruction of the speech waveform even if predicted without error. To address these issues and to make the excitation model more robust against errors in acoustic modeling, this paper proposes two modifications to the excitation model training scheme. First, the excitation model is trained in a connected manner, with inputs generated by the acoustic model. Second, the target glottal waveforms are re-estimated by performing glottal inverse filtering with the predicted vocal tract filters. The results show that both of these modifications improve performance measured in MSE and MFCC distortion, and slightly improve the subjective quality of the synthetic speech.

Linear prediction (LP) is a prevalent source-filter separation method of speech production. One of the drawbacks of conventional LP-based approaches is the biasing of estimated formants by harmonic peaks. Methods such as discrete all-pole modeling and weighted LP have been proposed to overcome this problem, but they all use a linear frequency scale. This study proposes a new LP technique, frequency-warped time-weighted linear prediction (WWLP), to provide spectral envelope estimates robust to harmonic peaks that work on a warped frequency scale that approximates the sensitivities of the human auditory system. Experiments are performed within the context of vocoding in statistical parametric speech synthesis. Subjective listening test results show that WWLP-based spectral envelope modeling increases quality over previously developed methods.

Achieving high quality and naturalness in statistical parametric synthesis of female voices remains to be difficult despite recent advances in the study area. Vocoding is one such key element in all statistical speech synthesizers that is known to affect the synthesis quality and naturalness. The present study focuses on a special type of vocoding, glottal vocoders, which aim to parameterize speech based on modelling the real excitation of (voiced) speech, the glottal flow. More specifically, we compare three different glottal vocoders by aiming at improved synthesis naturalness of female voices. Two of the vocoders are previously known, both utilizing an old glottal inverse filtering (GIF) method in estimating the glottal flow. The third on, denoted as Quasi Closed Phase - Deep Neural Net (QCP-DNN), takes advantage of a recently proposed new GIF method that shows improved accuracy in estimating the glottal flow from high-pitched speech. Subjective listening tests conducted on an US English female voice show that the proposed QCP-DNN method gives significant improvement in synthetic naturalness compared to the two previously developed glottal vocoders.

GlottHMM is a previously developed vocoder that has been successfully used in HMM-based synthesis by parameterizing speech into two parts (glottal flow, vocal tract) according to the functioning of the real human voice production mechanism. In this study, a new glottal vocoding method, GlottDNN, is proposed. The GlottDNN vocoder is built on the principles of its predecessor, GlottHMM, but the new vocoder introduces three main improvements: GlottDNN (1) takes advantage of a new, more accurate glottal inverse filtering method, (2) uses a new method of deep neural network (DNN) -based glottal excitation generation, and (3) proposes a new approach of band-wise processing of full-band speech. The proposed GlottDNN vocoder was evaluated as part of a full-band state-of-the-art DNN-based text-to-speech (TTS) synthesis system, and compared against the release version of the original GlottHMM vocoder, and the well-known STRAIGHT vocoder. The results of the subjective listening test indicate that GlottDNN improves the TTS quality over the compared methods.

In this paper, we describe a project that explores a novel experimental setup towards building a spoken, multi-modally rich, and human-like multiparty tutoring robot. A human-robot interaction setup is designed, and a human-human dialogue corpus is collected. The corpus targets the development of a dialogue system platform to study verbal and nonverbal tutoring strategies in multiparty spoken interactions with robots which are capable of spoken dialogue. The dialogue task is centered on two participants involved in a dialogue aiming to solve a card-ordering game. Along with the participants sits a tutor (robot) that helps the participants perform the task, and organizes and balances their interaction. Different multimodal signals captured and auto-synchronized by different audio-visual capture technologies, such as a microphone array, Kinects, and video cameras, were coupled with manual annotations. These are used build a situated model of the interaction based on the participants personalities, their state of attention, their conversational engagement and verbal dominance, and how that is correlated with the verbal and visual feed-back, turn-management, and conversation regulatory actions generated by the tutor. Driven by the analysis of the corpus, we will show also the detailed design methodologies for an affective, and multimodally rich dialogue system that allows the robot to measure incrementally the attention states, and the dominance for each participant, allowing the robot head Furhat to maintain a well-coordinated, balanced, and engaging conversation, that attempts to maximize the agreement and the contribution to solve the task.

This paper presents an experimental comparison of various leading vocoders for the application of HMM-based laughter synthesis. Four vocoders, commonly used in HMM-based speech synthesis, are used in copy-synthesis and HMM-based synthesis of both male and female laughter. Subjective evaluations are conducted to assess the performance of the vocoders. The results show that all vocoders perform relatively well in copy-synthesis. In HMM-based laughter synthesis using original phonetic transcriptions, all synthesized laughter voices were significantly lower in quality than copy-synthesis, indicating a challenging task and room for improvements. Interestingly, two vocoders using rather simple and robust excitation modeling performed the best, indicating that robustness in speech parameter extraction and simple parameter representation in statistical modeling are key factors in successful laughter synthesis.

In this paper, the effect of MPEG audio compression on HMMbased speech synthesis is studied. Speech signals are encoded with various compression rates and analyzed using the GlottHMM vocoder. Objec ...

In this paper, we propose modeling a noisy-channel for the task of voice conversion (VC). We have used the artificial neural networks (ANN) to capture speaker-specific characteristics of a target speaker which avoid the need for any training utterance from a source speaker. We use articulatory features (AFs) as a canonical form or speaker-independent representation of a speech signal. Our studies show that AFs contain a significant amount of speaker information in their trajectories. Suitable techniques are proposed to normalize the speaker-specific information in AF trajectories and the resultant AFs are used in voice conversion. The results of voice conversion evaluated using objective and subjective measures confirm that AFs can be used as a canonical form in nosiy-channel to capture speakerspecific characteristics of a target speaker.