In recent text-to-speech synthesis and voice conversion systems, a mel-spectrogram is commonly applied as an intermediate representation, and the necessity for a mel-spectrogram vocoder is ...increasing. A mel-spectrogram vocoder must solve three inverse problems: recovery of the original-scale magnitude spectrogram, phase reconstruction, and frequency-to-time conversion. A typical convolutional mel-spectrogram vocoder solves these problems jointly and implicitly using a convolutional neural network, including temporal upsampling layers, when directly calculating a raw waveform. Such an approach allows skipping redundant processes during waveform synthesis (e.g., the direct reconstruction of high-dimensional original-scale spectrograms). By contrast, the approach solves all problems in a black box and cannot effectively employ the time-frequency structures existing in a mel-spectrogram. We thus propose iSTFTNet, which replaces some output-side layers of the mel-spectrogram vocoder with the inverse short-time Fourier transform (iSTFT) after sufficiently reducing the frequency dimension using upsampling layers, reducing the computational cost from black-box modeling and avoiding redundant estimations of high-dimensional spectrograms. During our experiments, we applied our ideas to three HiFi-GAN variants and made the models faster and more lightweight with a reasonable speech quality. 1
This article proposes a voice conversion (VC) method using sequence-to-sequence (seq2seq or S2S) learning, which flexibly converts not only the voice characteristics but also the pitch contour and ...duration of input speech. The proposed method, called ConvS2S-VC, has three key features. First, it uses a model with a fully convolutional architecture. This is particularly advantageous in that it is suitable for parallel computations using GPUs. It is also beneficial since it enables effective normalization techniques such as batch normalization to be used for all the hidden layers in the networks. Second, it achieves many-to-many conversion by simultaneously learning mappings among multiple speakers using only a single model instead of separately learning mappings between each speaker pair using a different model. This enables the model to fully utilize available training data collected from multiple speakers by capturing common latent features that can be shared across different speakers. Owing to this structure, our model works reasonably well even without source speaker information, thus making it able to handle any-to-many conversion tasks. Third, we introduce a mechanism, called the conditional batch normalization that switches batch normalization layers in accordance with the target speaker. This particular mechanism has been found to be extremely effective for our many-to-many conversion model. We conducted speaker identity conversion experiments and found that ConvS2S-VC obtained higher sound quality and speaker similarity than baseline methods. We also found from audio examples that it could perform well in various tasks including emotional expression conversion, electrolaryngeal speech enhancement, and English accent conversion.
This paper is concerned with non-parallel whisper-to-normal speaking-style conversion (W2N-SC), which converts whispered speech into normal speech without using parallel training data. Most relevant ...to this task is voice conversion (VC), which converts one speaker's voice to another. However, the W2N-SC task differs from the regular VC task in three main respects. First, unlike normal speech, whispered speech contains little or no pitch information. Second, whispered speech usually has significantly less energy than normal speech and is therefore more susceptible to external noise. Third, in the actual usage scenario of W2N-SC, users may suddenly switch voice modes from whispered to normal speech, or vice versa, meaning that the speaking-style of input speech cannot be assumed in advance. To clarify whether existing VC techniques can successfully handle these task-specific concerns and how they should be modified to better address them, we consider a variational autoencoder (VAE)-based VC method as a baseline and examine what modifications to this method would be effective for the current task. Specifically, we study the effects of 1) a self-supervised training scheme called filling-in-frames (FIF); 2) data augmentation (DA) using noisy speech samples; and 3) an architecture that allows for any-to-many conversions. Through experimental evaluation of the W2N-SC and speaker conversion tasks, we confirmed that, especially in the W2N-SC task, the version incorporating the above modifications works better than the baseline VC model applied as is.
Facile and scalable fabrication of α-Fe2O3 photoanodes using a precursor solution containing FeIII ions and 1-ethylimidazole (EIm) in methanol was demonstrated to afford a rigidly adhered α-Fe2O3 ...film with a controllable thickness on a fluorine-doped tin oxide (FTO) substrate. EIm ligation to FeIII ions in the precursor solution brought about high crystallinity of three-dimensionally well-interconnected nanoparticles of α-Fe2O3 upon sintering. This is responsible for the 13.6 times higher photocurrent density (at 1.23 V vs reference hydrogen electrode (RHE)) for photoelectrochemical (PEC) water oxidation on the α-Fe2O3 (w-α-Fe2O3) photoanode prepared with EIm compared with that (w/o-α-Fe2O3) prepared without EIm. The w-α-Fe2O3 photoanode provided the highest charge separation efficiency (ηsep) value of 27% among the state-of-the-art pristine α-Fe2O3 photoanodes, providing incident photon-to-current conversion efficiency (IPCE) of 13% at 420 nm and 1.23 V vs RHE. The superior ηsep for the w-α-Fe2O3 photoanode is attributed to the decreased recombination of the photogenerated charge carriers at the grain boundary between nanoparticles, in addition to the higher number of the catalytically active sites and the efficient bulk charge transport in the film, compared with w/o-α-Fe2O3.
A heterogeneous catalyst incorporating an inorganic ion cofactor for electrochemical water oxidation was exploited using a CoO(OH) nanoparticle layer-deposited electrode. The significant catalytic ...current for water oxidation was generated in a Na2B4O7 solution at pH 9.4 when applying 0.94 V versus Ag/AgCl in contrast to no catalytic current generation in the K2SO4 solution at the same pH. HB4O7 – and B4O7 2– ions were indicated to act as key cofactors for the induced catalytic activity of the CoO(OH) layer. The Na2B4O7 concentration dependence of the catalytic current was analyzed based on a Michaelis–Menten-type kinetics to provide an affinity constant of cofactors to the active sites, K m = 28 ± 3.6 mM, and the maximum catalytic current density, I max = 2.3 ± 0.13 mA cm–2. The I max value of HB4O7 – and B4O7 2– ions was 1.4 times higher than that (1.3 mA cm–2) for the previously reported case of CO3 2– ions. This could be explained by the shorter-range proton transfer from the active site to the proton-accepting cofactor because of the larger size and more flexible conformation of HB4O7 – and B4O7 2– ions compared with that of CO3 2– ions.
This paper presents a novel speaking aid system to help laryngectomees produce more naturally sounding electrolaryngeal (EL) speech. An electrolarynx is an external device to generate excitation ...signals, instead of vibration of the vocal folds. Although the conventional EL speech is quite intelligible, its naturalness suffers from the unnatural fundamental frequency (F0) patterns of the mechanically generated excitation signals. To improve the naturalness of EL speech, we have proposed EL speech enhancement methods using statistical F0 pattern prediction. In these methods, the original EL speech recorded by a microphone is presented from a loudspeaker after performing the speech enhancement. These methods are effective for some situation, such as telecommunication, but it is not suitable for face-to-face conversation because not only the enhanced EL speech but also the original EL speech is presented to listeners. In this paper, to develop an EL speech enhancement also effective for face-to-face conversation, we propose a method for directly controlling F0 patterns of the excitation signals to be generated from the electrolarynx using the statistical F0 prediction. To get an "actual feel” of the proposed system, we also implement a prototype system. By using the prototype system, we find latency issues caused by a real-time processing. To address these latency issues, we furthermore propose segmental continuous F0 pattern modeling and forthcoming F0 pattern modeling. With evaluations through simulation, we demonstrate that our proposed system is capable of effectively addressing the issues of latency and those of electrolarynx in term of the naturalness.
This paper presents an electrolaryngeal (EL) speech enhancement method capable of significantly improving naturalness of EL speech while causing no degradation in its intelligibility. An ...electrolarynx is an external device that artificially generates excitation sounds to enable laryngectomees to produce EL speech. Although proficient laryngectomees can produce quite intelligible EL speech, it sounds very unnatural due to the mechanical excitation produced by the device. Moreover, the excitation sounds produced by the device often leak outside, adding to EL speech as noise. To address these issues, there are mainly two conventional approached to EL speech enhancement through either noise reduction or statistical voice conversion (VC). The former approach usually causes no degradation in intelligibility but yields only small improvements in naturalness as the mechanical excitation sounds remain essentially unchanged. On the other hand, the latter approach significantly improves naturalness of EL speech using spectral and excitation parameters of natural voices converted from acoustic parameters of EL speech, but it usually causes degradation in intelligibility owing to errors in conversion. We propose a hybrid approach using a noise reduction method for enhancing spectral parameters and statistical voice conversion method for predicting excitation parameters. Moreover, we further modify the prediction process of the excitation parameters to improve its prediction accuracy and reduce adverse effects caused by unvoiced/voiced prediction errors. The experimental results demonstrate the proposed method yields significant improvements in naturalness compared with EL speech while keeping intelligibility high enough.
Non-parallel voice conversion (VC) is a technique for learning the mapping from source to target speech without relying on parallel data. This is an important task, but it has been challenging due to ...the disadvantages of the training conditions. Recently, CycleGAN-VC has provided a breakthrough and performed comparably to a parallel VC method without relying on any extra data, modules, or time alignment procedures. However, there is still a large gap between the real target and converted speech, and bridging this gap remains a challenge. To reduce the gap, we propose CycleGAN-VC2, which is an improved version of CycleGAN-VC incorporating three new techniques: an improved objective (two-step adversarial losses), improved generator (2-1-2D CNN), and improved discriminator (PatchGAN). We evaluated our method on a non-parallel VC task and analyzed the effect of each technique in detail. An objective evaluation showed that these techniques help bring the converted feature sequence closer to the target in terms of both global and local structures, which we assess by using Mel-cepstral distortion and modulation spectra distance, respectively. A subjective evaluation showed that CycleGAN-VC2 outperforms CycleGAN-VC in terms of naturalness and similarity for every speaker pair, including intra-gender and inter-gender pairs. 1