Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2209.10890

EPIC TTS Models: Empirical Pruning Investigations Characterizing Text-To-Speech Models

Perry Lam·Huayun Zhang·Nancy F. Chen·Berrak Sisman
View original source ↗Read source document ↗Export metadata ↓

Abstract

Neural models are known to be over-parameterized, and recent work has shown that sparse text-to-speech (TTS) models can outperform dense models. Although a plethora of sparse methods has been proposed for other domains, such methods have rarely been applied in TTS. In this work, we seek to answer the question: what are the characteristics of selected sparse techniques on the performance and model complexity? We compare a Tacotron2 baseline and the results of applying five techniques. We then evaluate the performance via the factors of naturalness, intelligibility and prosody, while reporting model size and training time. Complementary to prior research, we find that pruning before or during training can achieve similar performance to pruning after training and can be trained much faster, while removing entire neurons degrades performance much more than removing parameters. To our best knowledge, this is the first work that compares sparsity paradigms in text-to-speech synthesis.

Explore related subjects

eess.AScs.LG

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Perry Lam, Huayun Zhang, Nancy F. Chen, Berrak Sisman. 2022-09-22. EPIC TTS Models: Empirical Pruning Investigations Characterizing Text-To-Speech Models. https://doi.org/10.21437/interspeech.2022-10626

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Cyclic MPDR Beamforming for Suppression of Almost-Cyclostationary Acoustic Interference

Conventional acoustic beamformers typically assume short-time stationarity and process frequency bins independently, ignoring inter-frequency correlations. This is suboptimal for almost-periodic noise sources such as engines, fans, and musical instruments: these signals are better modeled as (almost) cyclostationary (ACS) processes with statistically correlated spectral components. This paper introduces the cyclic minimum power distortionless response (cMPDR) beamformer, which extends the conventional MPDR to jointly exploit spatial and spectral correlations. Building on frequency-shifted (FRESH) filtering, it suppresses noise components that are coherent across harmonically related frequencies, reducing residual noise beyond what spatial filtering alone can achieve. To address inharmonicity, where partials deviate from exact integer multiples of a fundamental frequency, we estimate resonant frequencies from a periodogram and derive frequency shifts from their pairwise spacing. Theoretical analysis yields closed-form expressions for residual noise and proves that output power decreases monotonically with the number of cyclic components. Experiments on synthetic harmonic noise and real UAV motor recordings confirm these findings: in low-SNR scenarios, the cMPDR achieves up to 5 dB improvement in SI-SDR over the MPDR, yields consistent STOI gains, and remains effective with a single microphone. When spectral correlation is absent, the method reduces to conventional MPDR and does not degrade performance. These results suggest that cyclic processing is a viable direction for acoustic noise reduction that deserves further investigation. Code and audio samples are available at https://github.com/Screeen/cMPDR.

eess.AS↗
arXiv2026

Activation Steering for Accent-Neutralized Zero-Shot Text-To-Speech

Zero-shot Text-to-Speech (TTS) models can generate speech that captures both the voice timbre and accent of a reference speaker. However, disentangling these attributes and individually controlling the output accent remains challenging. In this study, we introduce a post-hoc and training-free approach to neutralize accent while preserving the speaker's original timbre, utilizing inference-time activation steering. We first extract layer-specific "steering vectors" offline, which are derived from the activation differences within the TTS model between accented and native speech. During inference, the steering vectors guide the model to produce accent-neutralized, timbre-preserving speech. Empirical results demonstrate that the proposed steering vectors effectively mitigate the output accent and exhibit strong generalizability to unseen accented speakers, offering a practical solution for accent-free voice cloning. We further confirm that speaker timbre is largely preserved via speaker embedding cosine similarity and speaker verification acceptance rate computed in an estimated accent-orthogonal subspace, as well as subjective listening tests. Audio demo: https://accentsteer.github.io/

eess.AS↗
arXiv2026

Active noise cancellation on open-ear smart glasses

Active noise cancellation (ANC) is widely deployed on consumer headphones and earbuds to suppress environmental noise. However, existing ANC systems require an error microphone at the user's ear canal to measure residual sound, preventing deployment on emerging open-ear wearable devices such as smart glasses and VR headsets, which leave the ear unoccluded. Here we present an ANC system for open-ear wearables that suppresses environmental noise using only microphones and miniaturized open-ear speakers embedded within the frame of the wearables, removing the need for an in-ear error microphone. Our low-latency computational pipeline uses a neural network to estimate the noise at the ear from an array of eight microphones distributed around the wearable's frame and generates an anti-noise signal in real-time. This mapping generalizes to unseen users and acoustic environments without prior acoustic measurement. We develop a custom glasses prototype and evaluate across eleven unseen users and eight unseen environments under mobility in the 100 to 1000 Hz frequency range, where environmental noise is concentrated. We achieve a mean noise reduction of 9.6 dB without any calibration, and 11.2 dB with a brief user-specific calibration. Further, we demonstrate that our approach extends to the broader class of open-ear wearables including VR headsets and headbands.

eess.AS↗