Clova Baseline System for the VoxCeleb Speaker Recognition Challenge 2020
By Hee-Soo Heo, Bong-Jin Lee, Jaesung Huh, Joon Son Chung. 2020
156 citations recorded by the source corpus.
This report describes our submission to the VoxCeleb Speaker Recognition Challenge (VoxSRC) at Interspeech 2020. We perform a careful analysis of speaker recognition models based on the popular ResNet architecture, and train a number of variants using a range of loss functions. Our results show significant improvements over most existing works without the use of model ensemble or post-processing. We release the training code and pre-trained models as unofficial baselines for this year's challenge.
Subfields: Artificial Intelligence, 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
- Moacir Antonelli Ponti
- Edresson Casanova
- Eren Gölge
- Julian Weber
- Arnaldo Cândido Júnior
- Christopher Dane Shulby
- Nima Mesgarani
- Cong Han
In Signal Processing
- Moacir Antonelli Ponti
- Edresson Casanova
- Eren Gölge
- Julian Weber
- Arnaldo Cândido Júnior
- Christopher Dane Shulby
Citations in the corpus
- YourTTS: Towards Zero-Shot Multi-Speaker TTS and Zero-Shot Voice Conversion for everyone (2021, positive): “As speaker encoder, the H/ASP model [27] is trained with the Prototypical Angular [28] plus Softmax loss functions in the VoxCeleb 2 [29] dataset.”
- XTTS: a Massively Multilingual Zero-Shot Text-to-Speech Model (2024, methodology): “Our proposed decoder is also conditioned with speaker embedding from the H/ASP model [24].”