Learning Robust Latent Representations for Controllable Speech Synthesis

• URL: http://arxiv.org/abs/2105.04458v1
• Date: Mon, 10 May 2021 15:49:03 GMT
• Title: Learning Robust Latent Representations for Controllable Speech Synthesis
• Authors: Shakti Kumar, Jithin Pradeep, Hussain Zaidi
• Abstract summary: We propose RTI-VAE (Reordered Transformer with Information reduction VAE) to minimize the mutual information between different latent variables. We show that RTI-VAE reduces the cluster overlap of speaker attributes by at least 30% over LSTM-VAE and by at least 7% over vanilla Transformer-VAE.
• Score: 0.0
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: State-of-the-art Variational Auto-Encoders (VAEs) for learning disentangled latent representations give impressive results in discovering features like pitch, pause duration, and accent in speech data, leading to highly controllable text-to-speech (TTS) synthesis. However, these LSTM-based VAEs fail to learn latent clusters of speaker attributes when trained on either limited or noisy datasets. Further, different latent variables start encoding the same features, limiting the control and expressiveness during speech synthesis. To resolve these issues, we propose RTI-VAE (Reordered Transformer with Information reduction VAE) where we minimize the mutual information between different latent variables and devise a modified Transformer architecture with layer reordering to learn controllable latent representations in speech data. We show that RTI-VAE reduces the cluster overlap of speaker attributes by at least 30\% over LSTM-VAE and by at least 7\% over vanilla Transformer-VAE.

Related papers

• Large Language Models Are Strong Audio-Visual Speech Recognition Learners [53.142635674428874]
  Multimodal large language models (MLLMs) have recently become a focal point of research due to their formidable multimodal understanding capabilities. We propose Llama-AVSR, a new MLLM with strong audio-visual speech recognition capabilities. We evaluate our proposed approach on LRS3, the largest public AVSR benchmark, and we achieve new state-of-the-art results for the tasks of ASR and AVSR with a WER of 0.81% and 0.77%, respectively.
  arXiv  Detail & Related papers  (2024-09-18T21:17:27Z)
• VQ-CTAP: Cross-Modal Fine-Grained Sequence Representation Learning for Speech Processing [81.32613443072441]
  For tasks such as text-to-speech (TTS), voice conversion (VC), and automatic speech recognition (ASR), a cross-modal fine-grained (frame-level) sequence representation is desired. We propose a method called Quantized Contrastive Token-Acoustic Pre-training (VQ-CTAP), which uses the cross-modal sequence transcoder to bring text and speech into a joint space.
  arXiv  Detail & Related papers  (2024-08-11T12:24:23Z)
• Prompt Tuning of Deep Neural Networks for Speaker-adaptive Visual Speech Recognition [66.94463981654216]
  We propose prompt tuning methods of Deep Neural Networks (DNNs) for speaker-adaptive Visual Speech Recognition (VSR) We finetune prompts on adaptation data of target speakers instead of modifying the pre-trained model parameters. The effectiveness of the proposed method is evaluated on both word- and sentence-level VSR databases.
  arXiv  Detail & Related papers  (2023-02-16T06:01:31Z)
• Cross-Speaker Emotion Transfer for Low-Resource Text-to-Speech Using Non-Parallel Voice Conversion with Pitch-Shift Data Augmentation [19.807274303199755]
  We propose a novel data augmentation method that combines pitch-shifting and VC techniques. Because pitch-shift data augmentation enables the coverage of a variety of pitch dynamics, it greatly stabilizes training for both VC and TTS models. Subjective test results showed that a FastSpeech 2-based emotional TTS system with the proposed method improved naturalness and emotional similarity compared with conventional methods.
  arXiv  Detail & Related papers  (2022-04-21T11:03:37Z)
• Robust Disentangled Variational Speech Representation Learning for Zero-shot Voice Conversion [34.139871476234205]
  We investigate zero-shot voice conversion from a novel perspective of self-supervised disentangled speech representation learning. A zero-shot voice conversion is performed by feeding an arbitrary speaker embedding and content embeddings to a sequential variational autoencoder (VAE) decoder. On TIMIT and VCTK datasets, we achieve state-of-the-art performance on both objective evaluation, i.e., speaker verification (SV) on speaker embedding and content embedding, and subjective evaluation, i.e. voice naturalness and similarity, and remains to be robust even with noisy source/target utterances.
  arXiv  Detail & Related papers  (2022-03-30T23:03:19Z)
• Streaming Multi-Talker ASR with Token-Level Serialized Output Training [53.11450530896623]
  t-SOT is a novel framework for streaming multi-talker automatic speech recognition. The t-SOT model has the advantages of less inference cost and a simpler model architecture. For non-overlapping speech, the t-SOT model is on par with a single-talker ASR model in terms of both accuracy and computational cost.
  arXiv  Detail & Related papers  (2022-02-02T01:27:21Z)
• Switching Variational Auto-Encoders for Noise-Agnostic Audio-visual Speech Enhancement [26.596930749375474]
  We introduce the use of a latent sequential variable with Markovian dependencies to switch between different VAE architectures through time. We derive the corresponding variational expectation-maximization algorithm to estimate the parameters of the model and enhance the speech signal.
  arXiv  Detail & Related papers  (2021-02-08T11:45:02Z)
• Pretraining Techniques for Sequence-to-Sequence Voice Conversion [57.65753150356411]
  Sequence-to-sequence (seq2seq) voice conversion (VC) models are attractive owing to their ability to convert prosody. We propose to transfer knowledge from other speech processing tasks where large-scale corpora are easily available, typically text-to-speech (TTS) and automatic speech recognition (ASR) We argue that VC models with such pretrained ASR or TTS model parameters can generate effective hidden representations for high-fidelity, highly intelligible converted speech.
  arXiv  Detail & Related papers  (2020-08-07T11:02:07Z)
• Should we hard-code the recurrence concept or learn it instead ? Exploring the Transformer architecture for Audio-Visual Speech Recognition [10.74796391075403]
  We present a variant of AV Align where the recurrent Long Short-term Memory (LSTM) block is replaced by the more recently proposed Transformer block. We find that Transformers also learn cross-modal monotonic alignments, but suffer from the same visual convergence problems as the LSTM model.
  arXiv  Detail & Related papers  (2020-05-19T09:06:39Z)
• Many-to-Many Voice Transformer Network [55.17770019619078]
  This paper proposes a voice conversion (VC) method based on a sequence-to-sequence (S2S) learning framework. It enables simultaneous conversion of the voice characteristics, pitch contour, and duration of input speech.
  arXiv  Detail & Related papers  (2020-05-18T04:02:08Z)
• Multiresolution and Multimodal Speech Recognition with Transformers [22.995102995029576]
  This paper presents an audio visual automatic speech recognition (AV-ASR) system using a Transformer-based architecture. We focus on the scene context provided by the visual information, to ground the ASR. Our results are comparable to state-of-the-art Listen, Attend and Spell-based architectures.
  arXiv  Detail & Related papers  (2020-04-29T09:32:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.