본문 바로가기

연구실 공부

[논문] BW-EDA-EEND: Streaming End-to-End Neural Speaker Diarization for a Variable Number of Speakers https://arxiv.org/abs/2011.02678 BW-EDA-EEND: Streaming End-to-End Neural Speaker Diarization for a Variable Number of Speakers We present a novel online end-to-end neural diarization system, BW-EDA-EEND, that processes data incrementally for a variable number of speakers. The system is based on the Encoder-Decoder-Attractor (EDA) architecture of Horiguchi et al., but utilizes the arxiv.org 해당 논.. 연구실 공부 2024. 3. 26.
[논문] End-to-End Speaker Diarization for an Unknown Number of Speakers with Encoder-Decoder Based Attractors https://arxiv.org/abs/2005.09921 End-to-End Speaker Diarization for an Unknown Number of Speakers with Encoder-Decoder Based Attractors End-to-end speaker diarization for an unknown number of speakers is addressed in this paper. Recently proposed end-to-end speaker diarization outperformed conventional clustering-based speaker diarization, but it has one drawback: it is less flexible in te arxiv.. 연구실 공부 2024. 3. 21.
[논문] End-to-End Neural Speaker Diarization with Self-Attention https://ieeexplore.ieee.org/abstract/document/9003959 Request Rejected ieeexplore.ieee.org 해당 논문을 보고 작성했습니다. Abstract Speaker diarization은 speaker embedding을 clustering하는 방식으로 주로 발전되어 왔습니다. 그러나, clustering 기반 방식들은 2가지 문제가 존재합니다. (1) diarization error를 직접적으로 minimize하는 방식으로 최적화할 수 없습니다. (2) speaker overlap을 정확하게 다룰 수 없습니다. 이러한 문제를 해결하기 위해, bidirectional long short-term memory (BLSTM) network를 사용하.. 연구실 공부 2024. 3. 18.
[논문] Integrating End-to-End Neural and Clustering-based Diarization: Getting the Best of Both Worlds https://arxiv.org/abs/2010.13366 Integrating end-to-end neural and clustering-based diarization: Getting the best of both worlds Recent diarization technologies can be categorized into two approaches, i.e., clustering and end-to-end neural approaches, which have different pros and cons. The clustering-based approaches assign speaker labels to speech regions by clustering speaker emb arxiv.org 해당.. 연구실 공부 2024. 3. 17.
[논문] Speaker Diarization Using Deep Neural Network Embeddings https://ieeexplore.ieee.org/document/7953094 Speaker diarization using deep neural network embeddings Speaker diarization is an important front-end for many speech technologies in the presence of multiple speakers, but current methods that employ i-vector clustering for short segments of speech are potentially too cumbersome and costly for the front-end ro ieeexplore.ieee.org 해당 논문을 보고 작성했습니다. A.. 연구실 공부 2024. 3. 14.
[논문] End-to-End Speaker Diarization as Post-Processing https://arxiv.org/abs/2012.10055 End-to-End Speaker Diarization as Post-Processing This paper investigates the utilization of an end-to-end diarization model as post-processing of conventional clustering-based diarization. Clustering-based diarization methods partition frames into clusters of the number of speakers; thus, they typically arxiv.org 해당 논문을 보고 작성했습니다. Abstract 이 논문에서 저자들은 전통적인 clust.. 연구실 공부 2024. 3. 14.
[survey 논문] A Review of Speaker Diarization: Recent Advances with Deep Learning https://arxiv.org/abs/2101.09624 A Review of Speaker Diarization: Recent Advances with Deep Learning Speaker diarization is a task to label audio or video recordings with classes that correspond to speaker identity, or in short, a task to identify "who spoke when". In the early years, speaker diarization algorithms were developed for speech recognition on arxiv.org 해당 survey paper를 보고 작성했습니다. Ab.. 연구실 공부 2024. 3. 13.
Speaker Diarization Speaker Diarization(화자 분리)에 대해 간단히 알아보겠습니다. 여러 명의 speaker가 말하고 있는 audio가 있다고 생각하겠습니다. 이때, audio file에서 '누가 언제 말했는지'의 문제를 다루는 것이 speaker diarization입니다. raw waveform이나 mel-spectrogram처럼 변환된 signal을 input으로 받아 '누가 언제 말했는지(who speak when)'에 대한 timestamp를 출력하는 방식입니다. speaker diarization은 3가지 순서로 진행됩니다. VAD & Segmentation VAD(Voice Activity Detection)은 음성 활동 감지입니다. 이는 음향 이벤트 감지(Sound Event Detection).. 연구실 공부 2024. 3. 11.
728x90

티스토리툴바