The Speakers in the Wild (SITW) Speaker Recognition Database

By Mitchell McLaren, Luciana Ferrer, Diego Castán, A. Lawson. 2016

279 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Evaluation & Benchmarks, Speech & Audio, 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

1. Andrew Zisserman
2. Joon Son Chung
3. Arsha Nagrani
4. Weidi Xie
5. Jaesung Huh
6. Samuel Albanie
7. Andrea Vedaldi
8. Seongkyu Mun

In Evaluation & Benchmarks

1. Andrew Zisserman
2. Arsha Nagrani
3. Joon Son Chung
4. Jaesung Huh
5. Samuel Albanie
6. Weidi Xie
7. Hee-Soo Heo
8. Bong-Jin Lee

In Speech & Audio

1. Andrew Zisserman
2. Joon Son Chung
3. Arsha Nagrani
4. Jaesung Huh
5. Weidi Xie
6. Samuel Albanie
7. Seongkyu Mun
8. Hee-Soo Heo

In Signal Processing

1. Andrew Zisserman
2. Arsha Nagrani
3. Joon Son Chung
4. Samuel Albanie
5. Jaesung Huh
6. Weidi Xie
7. Hee-Soo Heo
8. Bong-Jin Lee

Citations in the corpus

• VOXLINGUA107: A Dataset for Spoken Language Recognition (2020, background): “The Speakers in the Wild (SITW) database [6] contains speech samples of nearly 300 well-known public figures from open-source media.”
• VoxCeleb2: Deep Speaker Recognition (2018, background): “VoxCeleb1 [7] and SITW [8] are valuable contributions, however they are still an order of magnitude smaller than popular face datasets, which contain millions of images.”
• VoxCeleb: A Large-Scale Speaker Identification Dataset (2017, methodology): “Unlike the SITW dataset, both audio and video for each speaker is released.”
• Teacher-Student Training for Text-Independent Speaker Recognition (2018, methodology): “The SITWEVAL test set [30] comprises 180 speakers and is 12.”
• Voxceleb: Large-scale speaker verification in the wild (2020, methodology): “These are common metrics used by existing datasets and challenges, such as NIST SRE12 Greenberg (2012) and SITWMcLaren et al. (2016).”
• In defence of metric learning for speaker recognition (2020, background): “Large-scale datasets for speaker recognition such as the VoxCeleb [1, 2] and Speakers in the Wild [3] have become freely available, facilitating fast progress in the field.”
• Learnable PINs: Cross-Modal Embeddings for Person Identity (2018, background): “A similar path for generating a voice embedding is followed in the audio community for speaker recognition [27, 29].”
• Utterance-level Aggregation for Speaker Recognition in the Wild (2019, background): “Speaker recognition ‘in the wild’ has received an increasing amount of interest recently due to the availability of free large-scale datasets [1, 2, 3], and the easy accessibility of deep learning frameworks [4, 5, 6].”