BareWave: Waveform-Native Flow-Matching Text-to-Speech

• URL: http://arxiv.org/abs/2606.09048v1
• Date: Mon, 08 Jun 2026 05:36:42 GMT
• Title: BareWave: Waveform-Native Flow-Matching Text-to-Speech
• Abstract summary: We present BareWave, a fully waveform-native framework for direct text-to-wave generation in flow-matching TTS.<n>We develop a direct text-to-wave training framework that combines training-time representation alignment, staged noise scheduling, and velocity-aware perceptual alignment.<n> Experiments on zero-shot voice cloning show that strong intelligibility, speaker similarity, and naturalness can be achieved under a fully waveform-native inference path.
• Score: 76.5390412686083
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Removing intermediate representations and separately trained decoding stages has become an important direction in generative modeling. In text-to-speech, however, high-quality systems are still commonly built through an intermediate acoustic representation before waveform synthesis. In this work, we present BareWave, a fully waveform-native framework for direct text-to-wave generation in flow-matching TTS. We consider this setting to raise three training challenges: raw-waveform modeling lacks a strong pretrained representational scaffold, different stages of training benefit from different noise schedules, and data-space perceptual objectives do not automatically share the temporal structure of the velocity-space flow objective. As a result, direct waveform training is hard to optimize efficiently, hard to push toward a strong final operating point with a fixed recipe, and hard to integrate effective perceptual refinement. Guided by this view, we develop a direct text-to-wave training framework that combines training-time representation alignment, staged noise scheduling, and velocity-aware perceptual alignment (VAPA), while preserving a single waveform-native inference path without pretrained components at test time. Experiments on zero-shot voice cloning show that strong intelligibility, speaker similarity, and naturalness can be achieved under a fully waveform-native inference path, supporting waveform-native flow-matching TTS as a practical direction. Project page with audio demos is available at https://barewave.github.io/.

Related papers

• PhysWave: Physics-Guided Latent Diffusion Models for Controllable Spatial Audio Generation [11.945977059126008]
  We present PhysWave, a physics-guided latent diffusion model for controllable text-to-FOA generation.<n> PhysWave unifies natural-language and trajectory control through a shared waypoint-caption representation.<n>Extensive results show that the proposed priors help PhysWave generate spatially consistent FOA audio while maintaining competitive audio quality.
  arXiv  Detail & Related papers  (2026-08-30T04:37:50Z)
• WaveFormer: Frequency-Time Decoupled Vision Modeling with Wave Equation [24.13944601660532]
  Vision modeling has advanced rapidly with Transformers, whose attention mechanisms capture visual dependencies but lack a principled account of how semantic information propagates spatially.<n>We revisit this problem from a wave-based perspective, treating feature maps as spatial signals whose evolution over an internal propagation time is governed by an underdamped wave equation.<n>We propose a family of WaveFormer models as drop-in replacements for standard ViTs and CNNs, achieving competitive accuracy across image classification, object detection, and semantic segmentation.
  arXiv  Detail & Related papers  (2026-01-13T14:47:22Z)
• Pseudo-Autoregressive Neural Codec Language Models for Efficient Zero-Shot Text-to-Speech Synthesis [64.12708207721276]
  We introduce a novel pseudo-autoregressive (PAR) language modeling approach that unifies AR and NAR modeling.<n>Building on PAR, we propose PALLE, a two-stage TTS system that leverages PAR for initial generation followed by NAR refinement.<n>Experiments demonstrate that PALLE, trained on LibriTTS, outperforms state-of-the-art systems trained on large-scale data.
  arXiv  Detail & Related papers  (2025-04-14T16:03:21Z)
• High-Fidelity Speech Synthesis with Minimal Supervision: All Using Diffusion Models [56.00939852727501]
  Minimally-supervised speech synthesis decouples TTS by combining two types of discrete speech representations. Non-autoregressive framework enhances controllability, and duration diffusion model enables diversified prosodic expression.
  arXiv  Detail & Related papers  (2023-09-27T09:27:03Z)
• Differentiable Duration Modeling for End-to-End Text-to-Speech [6.571447892202893]
  parallel text-to-speech (TTS) models have recently enabled fast and highly-natural speech synthesis. We propose a differentiable duration method for learning monotonic sequences between input and output. Our model learns to perform high-fidelity synthesis through a combination of adversarial training and matching the total ground-truth duration.
  arXiv  Detail & Related papers  (2022-03-21T15:14:44Z)
• Adaptation of Tacotron2-based Text-To-Speech for Articulatory-to-Acoustic Mapping using Ultrasound Tongue Imaging [48.7576911714538]
  This paper experiments with transfer learning and adaptation of a Tacotron2 text-to-speech model to improve articulatory-to-acoustic mapping. We use a multi-speaker pre-trained Tacotron2 TTS model and a pre-trained WaveGlow neural vocoder.
  arXiv  Detail & Related papers  (2021-07-26T09:19:20Z)
• WaveGrad 2: Iterative Refinement for Text-to-Speech Synthesis [80.60577805727624]
  WaveGrad 2 is a non-autoregressive generative model for text-to-speech synthesis. It can generate high fidelity audio, approaching the performance of a state-of-the-art neural TTS system.
  arXiv  Detail & Related papers  (2021-06-17T17:09:21Z)
• Wave-Tacotron: Spectrogram-free end-to-end text-to-speech synthesis [25.234945748885348]
  We describe a sequence-to-sequence neural network which directly generates speech waveforms from text inputs. The architecture extends the Tacotron model by incorporating a normalizing flow into the autoregressive decoder loop. Experiments show that the proposed model generates speech with quality approaching a state-of-the-art neural TTS system.
  arXiv  Detail & Related papers  (2020-11-06T19:30:07Z)
• End-to-End Adversarial Text-to-Speech [33.01223309795122]
  We learn to synthesise speech from normalised text or phonemes in an end-to-end manner. Our proposed generator is feed-forward and thus efficient for both training and inference. It learns to produce high fidelity audio through a combination of adversarial feedback and prediction losses.
  arXiv  Detail & Related papers  (2020-06-05T17:41:05Z)
• Single Channel Speech Enhancement Using Temporal Convolutional Recurrent Neural Networks [23.88788382262305]
  temporal convolutional recurrent network (TCRN) is an end-to-end model that directly map noisy waveform to clean waveform. We show that our model is able to improve the performance of model, compared with existing convolutional recurrent networks.
  arXiv  Detail & Related papers  (2020-02-02T04:26:50Z)
• Temporal-Spatial Neural Filter: Direction Informed End-to-End Multi-channel Target Speech Separation [66.46123655365113]
  Target speech separation refers to extracting the target speaker's speech from mixed signals. Two main challenges are the complex acoustic environment and the real-time processing requirement. We propose a temporal-spatial neural filter, which directly estimates the target speech waveform from multi-speaker mixture.
  arXiv  Detail & Related papers  (2020-01-02T11:12:50Z)

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.