Automatic methods to predict listener opinions of synthesized speech remain elusive since listeners, systems being evaluated, characteristics of the speech, and even the instructions given and the rating scale all vary from test to test. While automatic predictors for metrics such as mean opinion score (MOS) can achieve high prediction accuracy on samples from the same test, they typically fail to generalize well to new listening test contexts. In this paper, using a variety of networks for MOS prediction including MOSNet and self-supervised speech models such as wav2vec2, we investigate their performance on data from different listening tests in both zero-shot and fine-tuned settings. We find that wav2vec2 models fine-tuned for MOS prediction have good generalization capability to out-of-domain data even for the most challenging case of utterance-level predictions in the zero-shot setting, and that fine-tuning to in-domain data can improve predictions. We also observe that unseen systems are especially challenging for MOS prediction models.
Are end-to-end text-to-speech (TTS) models over-parametrized? To what extent can these models be pruned, and what happens to their synthesis capabilities? This work serves as a starting point to explore pruning both spectrogram prediction networks and vocoders. We thoroughly investigate the tradeoffs between sparstiy and its subsequent effects on synthetic speech. Additionally, we explored several aspects of TTS pruning: amount of finetuning data versus sparsity, TTS-Augmentation to utilize unspoken text, and combining knowledge distillation and pruning. Our findings suggest that not only are end-to-end TTS models highly prunable, but also, perhaps surprisingly, pruned TTS models can produce synthetic speech with equal or higher naturalness and intelligibility, with similar prosody. All of our experiments are conducted on publicly available models, and findings in this work are backed by large-scale subjective tests and objective measures. Code and 200 pruned models are made available to facilitate future research on efficiency in TTS.
In this paper, we provide a series of multi-tasking benchmarks for simultaneously detecting spoofing at the segmental and utterance levels in the PartialSpoof database. First, we propose the SELCNN network, which inserts squeeze-and-excitation (SE) blocks into a light convolutional neural network (LCNN) to enhance the capacity of hidden feature selection. Then, we implement multi-task learning (MTL) frameworks with SELCNN followed by bidirectional long short-term memory (Bi-LSTM) as the basic model. We discuss MTL in PartialSpoof in terms of architecture (uni-branch/multi-branch) and training strategies (from-scratch/warm-up) step-by-step. Experiments show that the multi-task model performs better than single-task models. Also, in MTL, binary-branch architecture more adequately utilizes information from two levels than a uni-branch model. For the binary-branch architecture, fine-tuning a warm-up model works better than training from scratch. Models can handle both segment-level and utterance-level predictions simultaneously overall under binary-branch multi-task architecture. Furthermore, the multi-task model trained by fine-tuning a segmental warm-up model performs relatively better at both levels except on the evaluation set for segmental detection. Segmental detection should be explored further.
Timbre representations of musical instruments, essential for diverse applications such as musical audio synthesis and separation, might be learned as bottleneck features from an instrumental recognition model. Given the similarities between speaker recognition and musical instrument recognition, in this paper, we investigate how to adapt successful speaker recognition algorithms to musical instrument recognition to learn meaningful instrumental timbre representations. To address the mismatch between musical audio and models devised for speech, we introduce a group of trainable filters to generate proper acoustic features from input raw waveforms, making it easier for a model to be optimized in an input-agnostic and end-to-end manner. Through experiments on both the NSynth and RWC databases in both musical instrument closed-set identification and open-set verification scenarios, the modified speaker recognition model was capable of generating discriminative embeddings for instrument and instrument-family identities. We further conducted extensive experiments to characterize the encoded information in learned timbre embeddings.
Speech synthesis and music audio generation from symbolic input differ in many aspects but share some similarities. In this study, we investigate how text-to-speech synthesis techniques can be used for piano MIDI-to-audio synthesis tasks. Our investigation includes Tacotron and neural source-filter waveform models as the basic components, with which we build MIDI-to-audio synthesis systems in similar ways to TTS frameworks. We also include reference systems using conventional sound modeling techniques such as sample-based and physical-modeling-based methods. The subjective experimental results demonstrate that the investigated TTS components can be applied to piano MIDI-to-audio synthesis with minor modifications. The results also reveal the performance bottleneck -- while the waveform model can synthesize high quality piano sound given natural acoustic features, the conversion from MIDI to acoustic features is challenging. The full MIDI-to-audio synthesis system is still inferior to the sample-based or physical-modeling-based approaches, but we encourage TTS researchers to test their TTS models for this new task and improve the performance.
Shared challenges provide a venue for comparing systems trained on common data using a standardized evaluation, and they also provide an invaluable resource for researchers when the data and evaluation results are publicly released. The Blizzard Challenge and Voice Conversion Challenge are two such challenges for text-to-speech synthesis and for speaker conversion, respectively, and their publicly-available system samples and listening test results comprise a historical record of state-of-the-art synthesis methods over the years. In this paper, we revisit these past challenges and conduct a large-scale listening test with samples from many challenges combined. Our aims are to analyze and compare opinions of a large number of systems together, to determine whether and how opinions change over time, and to collect a large-scale dataset of a diverse variety of synthetic samples and their ratings for further research. We found strong correlations challenge by challenge at the system level between the original results and our new listening test. We also observed the importance of the choice of speaker on synthesis quality.
This work examines the content and usefulness of disentangled phone and speaker representations from two separately trained VQ-VAE systems: one trained on multilingual data and another trained on monolingual data. We explore the multi- and monolingual models using four small proof-of-concept tasks: copy-synthesis, voice transformation, linguistic code-switching, and content-based privacy masking. From these tasks, we reflect on how disentangled phone and speaker representations can be used to manipulate speech in a meaningful way. Our experiments demonstrate that the VQ representations are suitable for these tasks, including creating new voices by mixing speaker representations together. We also present our novel technique to conceal the content of targeted words within an utterance by manipulating phone VQ codes, while retaining speaker identity and intelligibility of surrounding words. Finally, we discuss recommendations for further increasing the viability of disentangled representations.