AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Losses Can Be Blessings: Routing Self-Supervised Speech Representations Towards Efficient Multilingual and Multitask Speech Processing

Yonggan Fu, Yang Zhang, Kaizhi Qian, Zhifan Ye, Zhongzhi Yu, Cheng-I Lai, Yingyan Lin

NeurIPS 2022 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Self-supervised speech representations · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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
ORI mask initialization instead of weight-magnitude (WMI) (ablation)S3-Router · LibriSpeech 10mWER test-clean: 40.09 -> 36.68Improved
random (RI) instead of ORI mask initialization (ablation)S3-Router · LibriSpeech 10mWER test-clean: 36.68 -> 36.98Hurt
random mask initialization compared with weight finetuning (ablation)S3-Router · LibriSpeech 10hWER test-clean: 9.7 -> 9.78No clear effect
finetune connections with S3-Router instead of weightswav2vec2-base · LibriSpeech 10mWER test-clean: 38.40 -> 36.68Improved
add a 4-gram LM decoderS3-Router vs weight finetuning · LibriSpeech 10mWER test-clean: 24.83 -> 23.17Improved
S3-Router instead of weight finetuning on data2vecdata2vec · LibriSpeech 10mWER test-clean: 30.75 -> 29.86Improved
S3-Router sparsity 0.10 instead of 0.09 on data2vec (ablation)S3-Router · LibriSpeech 10mWER test-clean: 29.86 -> 31.30Hurt
S3-Router instead of weight finetuning on 8 SUPERB taskswav2vec2-base · SUPERB speaker identificationAcc: 66.8 -> 71.07Improved
S3-Router instead of adaptor tuningwav2vec2-base · LibriSpeech 10mWER test-clean: 41.82 -> 36.68Improved
tune only FFN connections instead of self-attention only (ablation)S3-Router · LibriSpeech 1hWER test-clean: 24.39 -> 18.62Improved
tune both FFN and SA instead of FFN only (ablation)S3-Router · LibriSpeech 1hWER test-clean: 18.62 -> 19.26No clear effect
tune only the 3rd block group instead of the 1st (ablation)S3-Router · LibriSpeech 1hWER test-clean: 85.68 -> 24.57Improved

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