Most neural text-to-speech (TTS) models require paired data from the desired speaker for high-quality speech synthesis, which limits the usage of large amounts of untranscribed data for training. In this work, we present Guided-TTS, a high-quality TTS model that learns to generate speech from untranscribed speech data. Guided-TTS combines an unconditional diffusion probabilistic model with a separately trained phoneme classifier for text-to-speech. By modeling the unconditional distribution for speech, our model can utilize the untranscribed data for training. For text-to-speech synthesis, we guide the generative process of the unconditional DDPM via phoneme classification to produce mel-spectrograms from the conditional distribution given transcript. We show that Guided-TTS achieves comparable performance with the existing methods without any transcript for LJSpeech. Our results further show that a single speaker-dependent phoneme classifier trained on multispeaker large-scale data can guide unconditional DDPMs for various speakers to perform TTS.

We describe a neural network-based system for text-to-speech (TTS) synthesis that is able to generate speech audio in the voice of many different speakers, including those unseen during training. Our system consists of three independently trained components: (1) a speaker encoder network, trained on a speaker verification task using an independent dataset of noisy speech from thousands of speakers without transcripts, to generate a fixed-dimensional embedding vector from seconds of reference speech from a target speaker; (2) a sequence-to-sequence synthesis network based on Tacotron 2, which generates a mel spectrogram from text, conditioned on the speaker embedding; (3) an auto-regressive WaveNet-based vocoder that converts the mel spectrogram into a sequence of time domain waveform samples. We demonstrate that the proposed model is able to transfer the knowledge of speaker variability learned by the discriminatively-trained speaker encoder to the new task, and is able to synthesize natural speech from speakers that were not seen during training. We quantify the importance of training the speaker encoder on a large and diverse speaker set in order to obtain the best generalization performance. Finally, we show that randomly sampled speaker embeddings can be used to synthesize speech in the voice of novel speakers dissimilar from those used in training, indicating that the model has learned a high quality speaker representation.

We describe a neural network-based system for text-to-speech (TTS) synthesis that is able to generate speech audio in the voice of many different speakers, including those unseen during training. Our system consists of three independently trained components: (1) a speaker encoder network, trained on a speaker verification task using an independent dataset of noisy speech from thousands of speakers without transcripts, to generate a fixed-dimensional embedding vector from seconds of reference speech from a target speaker; (2) a sequence-to-sequence synthesis network based on Tacotron 2, which generates a mel spectrogram from text, conditioned on the speaker embedding; (3) an auto-regressive WaveNet-based vocoder that converts the mel spectrogram into a sequence of time domain waveform samples. We demonstrate that the proposed model is able to transfer the knowledge of speaker variability learned by the discriminatively-trained speaker encoder to the new task, and is able to synthesize natural speech from speakers that were not seen during training. We quantify the importance of training the speaker encoder on a large and diverse speaker set in order to obtain the best generalization performance. Finally, we show that randomly sampled speaker embeddings can be used to synthesize speech in the voice of novel speakers dissimilar from those used in training, indicating that the model has learned a high quality speaker representation.

Custom voice, a specific text to speech (TTS) service in commercial speech platforms, aims to adapt a source TTS model to synthesize personal voice for a target speaker using few speech data. Custom voice presents two unique challenges for TTS adaptation: 1) to support diverse customers, the adaptation model needs to handle diverse acoustic conditions that could be very different from source speech data, and 2) to support a large number of customers, the adaptation parameters need to be small enough for each target speaker to reduce memory usage while maintaining high voice quality. In this work, we propose AdaSpeech, an adaptive TTS system for high-quality and efficient customization of new voices. We design several techniques in AdaSpeech to address the two challenges in custom voice: 1) To handle different acoustic conditions, we use two acoustic encoders to extract an utterance-level vector and a sequence of phoneme-level vectors from the target speech during training; in inference, we extract the utterance-level vector from a reference speech and use an acoustic predictor to predict the phoneme-level vectors. 2) To better trade off the adaptation parameters and voice quality, we introduce conditional layer normalization in the mel-spectrogram decoder of AdaSpeech, and fine-tune this part in addition to speaker embedding for adaptation. We pre-train the source TTS model on LibriTTS datasets and fine-tune it on VCTK and LJSpeech datasets (with different acoustic conditions from LibriTTS) with few adaptation data, e.g., 20 sentences, about 1 minute speech. Experiment results show that AdaSpeech achieves much better adaptation quality than baseline methods, with only about 5K specific parameters for each speaker, which demonstrates its effectiveness for custom voice. Audio samples are available at https://speechresearch.github.io/adaspeech/.

In this paper, we propose an architecture to solve a novel problem statement that has stemmed more so in recent times with an increase in demand for virtual content delivery due to the COVID-19 pandemic. All educational institutions, workplaces, research centers, etc. are trying to bridge the gap of communication during these socially distanced times with the use of online content delivery. The trend now is to create presentations, and then subsequently deliver the same using various virtual meeting platforms. The time being spent in such creation of presentations and delivering is what we try to reduce and eliminate through this paper which aims to use Machine Learning (ML) algorithms and Natural Language Processing (NLP) modules to automate the process of creating a slides-based presentation from a document, and then use state-of-the-art voice cloning models to deliver the content in the desired author's voice. We consider a structured document such as a research paper to be the content that has to be presented. The research paper is first summarized using BERT summarization techniques and condensed into bullet points that go into the slides. Tacotron inspired architecture with Encoder, Synthesizer, and a Generative Adversarial Network (GAN) based vocoder, is used to convey the contents of the slides in the author's voice (or any customized voice). Almost all learning has now been shifted to online mode, and professionals are now working from the comfort of their homes. Due to the current situation, teachers and professionals have shifted to presentations to help them in imparting information. In this paper, we aim to reduce the considerable amount of time that is taken in creating a presentation by automating this process and subsequently delivering this presentation in a customized voice, using a content delivery mechanism that can clone any voice using a short audio clip.