DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | 양은호 | - |
dc.contributor.author | Kim, Changhun | - |
dc.contributor.author | 김창훈 | - |
dc.date.accessioned | 2024-07-30T19:30:41Z | - |
dc.date.available | 2024-07-30T19:30:41Z | - |
dc.date.issued | 2024 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=1096077&flag=dissertation | en_US |
dc.identifier.uri | http://hdl.handle.net/10203/321372 | - |
dc.description | 학위논문(석사) - 한국과학기술원 : 김재철AI대학원, 2024.2,[iv, 22 p. :] | - |
dc.description.abstract | In real-world scenarios, automatic speech recognition (ASR) models often encounter data distribution shifts, leading to erroneous predictions. To tackle this issue, a recent test-time adaptation (TTA) method has been proposed to adapt the pre-trained ASR model to the unlabeled target domain without source data. Despite decent performance gain, this approach relies solely on naive greedy decoding and performs adaptation across timesteps at the frame level, which may not be optimal given the sequential nature of model outputs. Motivated by this limitation, this thesis introduces a novel Sequential-level Generalized Entropy Minimization (SGEM) framework for general ASR models. To handle sequential output, SGEM first exploits beam search to explore candidate output logits and selects the most plausible one. Then, it utilizes generalized entropy minimization and negative sampling as effective unsupervised objectives to adapt the model. Through extensive experiments, SGEM verifies its state-of-the-art performance across three mainstream ASR models under various distribution shifts. | - |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | 기계 학습▼a음성 인식▼a분포 변화 강건성▼a테스트타임 적응▼a빔 서치▼a엔트로피 최소화▼a네거티브 샘플링 | - |
dc.subject | Machine learning▼aAutomatic speech recognition▼aDistribution shift robustness▼aTest-time adaptation▼aBeam search▼aEntropy minimization▼aNegative sampling | - |
dc.title | Test-time adaptation for automatic speech recognition via sequential-level generalized entropy minimization | - |
dc.title.alternative | 문장 수준의 일반화된 엔트로피 최소화를 통한 음성 인식 모델에 대한 테스트타임 적응 | - |
dc.type | Thesis(Master) | - |
dc.identifier.CNRN | 325007 | - |
dc.description.department | 한국과학기술원 :김재철AI대학원, | - |
dc.contributor.alternativeauthor | Yang, Eunho | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.