The Albayzin 2008 Language Recognition Evaluation

By Luis Javier Rodriguez-Fuentes, Mikel Peñagarikano, Amparo Varona, Mireia Díez, Germán Bordel. 2010

50 citations recorded by the source corpus.

Subfields: Evaluation & Benchmarks, Speech & Audio

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Evaluation & Benchmarks

1. Tanel Alumäe
2. Jörgen Valk
3. Seyed Omid Sadjadi

In Speech & Audio

1. Tanel Alumäe
2. Jörgen Valk
3. Mireia Díez
4. Seyed Omid Sadjadi

Citations in the corpus

• VOXLINGUA107: A Dataset for Spoken Language Recognition (2020, methodology): “…[3] consists of three partitions: training data, containing mostly broadcast speech for a 6 language subset (Basque, Catalan, English, Galician, Portugese and Spanish) [21], development data and test data, containing YouTube audio for 21 We experimented with several methods in this experiment.”
• The 2017 NIST Language Recognition Evaluation (2018, background): “The multiclass cross-entropy Hmce measures the information a language recognition system provides through the loglikelihood scores and is defined as follows [20]”
• KALAKA-3: a database for the assessment of spoken language recognition technology on YouTube audios (2015, methodology): “The training dataset comes entirely from KALAKA-2 (Rodriguez-Fuentes et al. 2012b), the database created to support the Albayzin 2010 LRE, which was an extension of KALAKA (Rodriguez-Fuentes et al. 2010b), the database that supported the Albayzin 2008 LRE.”