Deep Speaker Embeddings for Short-Duration Speaker Verification
By Gautam Bhattacharya, Md. Jahangir Alam, Patrick Kenny. 2017
150 citations recorded by the source corpus.
Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Andrew Zisserman
- Joon Son Chung
- Arsha Nagrani
- Weidi Xie
- Yoshua Bengio
- Mirco Ravanelli
- Andrea Vedaldi
- Junichi Yamagishi
In Computer Vision and Pattern Recognition
- Andrew Zisserman
- Arsha Nagrani
- Luc Van Gool
- Andrea Vedaldi
- Kristen Grauman
- Cordelia Schmid
- Xiaogang Wang
- Dacheng Tao
In Signal Processing
- Andrew Zisserman
- Joon Son Chung
- Arsha Nagrani
- Yoshua Bengio
- Weidi Xie
- Mirco Ravanelli
- Triantafyllos Afouras
- K. Shinoda
Citations in the corpus
- Exploring the Encoding Layer and Loss Function in End-to-End Speaker and Language Recognition System (2018, methodology): “We implement the SAP layer similar to [27, 28, 29].”
- Interpretable Convolutional Filters with SincNet (2018, background): “Most prior attempts exploit magnitude spectrogram features [22, 23, 24, 46, 47, 48].”
- Attentive Statistics Pooling for Deep Speaker Embedding (2018, methodology): “In fixed-phrase text-dependent speaker verification, an end-to-end neural network-based method has been proposed [6] in which Long Short-Term Memory (LSTM) with a single output from the last frame is used to obtain utterance-level speaker features, and it has out-performed conventional i-vector extrac”
- Speaker Recognition from Raw Waveform with SincNet (2018, background): “To mitigate this drawback, some recent works have proposed directly feeding the network with spectrogram bins [18–20] or even with raw waveforms [21–30].”
- Voxceleb: Large-scale speaker verification in the wild (2020, methodology): “The trunk DNN architecture used is often either a 2D CNN with convolutions in both the time and frequency domain (Nagrani et al., 2017; Chung et al., 2018; Cai et al., 2018c; 2018b; Hajibabaei and Dai, 2018; Bhattacharya et al., 2017), or a 1D CNN with convolutions applied only to the time domain (S”
- Utterance-level Aggregation for Speaker Recognition in the Wild (2019, methodology): “The trunk DNN architecture used is often either a 2D CNN with convolutions in both the time and frequency domain [2, 3, 15, 18, 19, 20], or a 1D CNN with convolutions applied only to the time domain [11, 12, 13, 21].”