AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Sample Efficient Adaptive Text-to-Speech

Yutian Chen, Yannis Assael, Brendan Shillingford, David Budden, Scott Reed, Heiga Zen, Quan Wang, Luis C. Cobo, Andrew Trask, Ben Laurie, Caglar Gulcehre, Aäron van den Oord and 2 more

ICLR 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Multilingual language representation pretraining · 104 citations (OpenAlex)

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Fine-tune the full WaveNet after fitting the speaker embedding, rather than fitting only the embeddingSEA-All versus SEA-Emb · LibriSpeech and VCTKnaturalness MOS: 3.86 ± 0.07 -> 3.94 ± 0.08; 3.95 ± 0.07 -> 4.13 ± 0.06; 3.81 ± 0.07 -> 3.92 ± 0.07; 3.82 ± 0.07 -> 3.92 ± 0.07Improved
Fine-tune the full WaveNet after fitting the speaker embedding, rather than fitting only the embeddingSEA-All versus SEA-Emb · VCTK, <10m adaptationvoice similarity MOS: 3.18 ± 0.10 -> 3.97 ± 0.09Improved
Use an audio-conditioned embedding encoder instead of fitting the embedding from adaptation dataSEA-Enc versus SEA-Emb · LibriSpeech and VCTKvoice similarity MOS: 3.42 ± 0.10 -> 2.47 ± 0.09; 3.56 ± 0.10 -> 2.59 ± 0.09; 3.07 ± 0.10 -> 2.07 ± 0.08; 3.18 ± 0.10 -> 2.19 ± 0.09Hurt
Increase adaptation data beyond 10 secondsSEA-All · LibriSpeechnaturalness MOS: 3.94 ± 0.08 -> 4.13 ± 0.06Improved
Increase adaptation data beyond 10 secondsSEA-All · VCTKnaturalness MOS: 3.92 ± 0.07 -> 3.92 ± 0.07No clear effect
Increase adaptation data from 10 seconds to <5mSEA-All · LibriSpeechspeaker verification EER: 3.17 -> 1.85Improved
Increase adaptation data from 10 seconds to <10mSEA-All · VCTKspeaker verification EER: 7.34 -> 4.33Improved
Use embedding-only fitting instead of the learned encoderSEA-Emb versus SEA-Enc · LibriSpeech and VCTKspeaker verification EER: 9.42 -> 2.74; 25.23 -> 10.24Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy