Towards Building Text-to-Speech Systems for the Next Billion Users

Karthik, Kumar, Gokul; Praveen, S V,; Kumar, Pratyush; Khapra, Mitesh M.; Nandakumar, Karthik

doi:10.1109/icassp49357.2023.10096069

Cited by 2 publications

(1 citation statement)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It may not be sensitive to some types of artifacts that affect speech quality, such as jitter, shimmer, and reverberation. MCD will not be able to capture the intelligibility of the speech as previously it has been observed that models having better MCD scores produce less intelligible speech [22].…”

Section: Results and Analysismentioning

confidence: 85%

SALTTS: Leveraging Self-Supervised Speech Representations for improved Text-to-Speech Synthesis

Sivaguru,

Lodagala,

Umesh

2023

Interspeech 2023

View full text Add to dashboard Cite

While FastSpeech2 aims to integrate aspects of speech such as pitch, energy, and duration as conditional inputs, it still leaves scope for richer representations. As a part of this work, we leverage representations from various Self-Supervised Learning (SSL) models to enhance the quality of the synthesized speech. In particular, we pass the FastSpeech2 encoder's length-regulated outputs through a series of encoder layers with the objective of reconstructing the SSL representations. In the SALTTS-parallel implementation, the representations from this second encoder are used for an auxiliary reconstruction loss with the SSL features. The SALTTS-cascade implementation, however, passes these representations through the decoder in addition to having the reconstruction loss. The richness of speech characteristics from the SSL features reflects in the output speech quality, with the objective and subjective evaluation measures of the proposed approach outperforming the baseline FastSpeech2.

show abstract

Section: Results and Analysismentioning

confidence: 85%