GLA-Grad++: An Improved Griffin-Lim Guided Diffusion Model for Speech Synthesis
This is an incremental improvement for speech synthesis systems that need to handle diverse audio inputs.
The paper tackles the problem of vocoder performance degradation when conditioning mel spectrograms diverge from the training distribution in speech synthesis, achieving consistent improvements over baseline models, especially in out-of-domain scenarios.
Recent advances in diffusion models have positioned them as powerful generative frameworks for speech synthesis, demonstrating substantial improvements in audio quality and stability. Nevertheless, their effectiveness in vocoders conditioned on mel spectrograms remains constrained, particularly when the conditioning diverges from the training distribution. The recently proposed GLA-Grad model introduced a phase-aware extension to the WaveGrad vocoder that integrated the Griffin-Lim algorithm (GLA) into the reverse process to reduce inconsistencies between generated signals and conditioning mel spectrogram. In this paper, we further improve GLA-Grad through an innovative choice in how to apply the correction. Particularly, we compute the correction term only once, with a single application of GLA, to accelerate the generation process. Experimental results demonstrate that our method consistently outperforms the baseline models, particularly in out-of-domain scenarios.