Blog(1)

• [INTERSPEECH 2024 Series #1] Relational Proxy Loss for Audio–Text based Keyword Spotting

  Keyword spotting (KWS) aims at identifying pre-specified keywords in audio streams. Recently, KWS has attracted considerable research attention with the popularity of voice assistants that are triggered by keywords such as “Alexa,” “Hi Bixby,” or “Okay Google.” KWS techniques can be broadly classified into two categories: fixed KWS [1-3] and flexible (or user-defined) KWS [4-13].

Research Areas(0)

Publications(6)

• Improved Small-Footprint ASR-Based Solution for Open Vocabulary Keyword Spotting

  AuthorMikolaj Pudo, Mateusz Wosik, Artur Janicki

  PublishedIEEE Access

  Date2024-07-01

• Joint Embedding Learning and Latent Subspace Probing for Cross-Domain Few-Shot Keyword Spotting

  AuthorMete Ozay

  PublishedInternational Conference on Acoustics, Speech and Signal Processing (ICASSP)

  Date2024-03-18

• Robust Keyword Spotting for Noisy Environments by Leveraging Speech Enhancement and Speech Presence Probability

  AuthorChouchang Yang,Yashas Malur Saidutta,Rakshith Srinivasa,Chinghua Lee,Yilin Shen,Hongxia Jin

  PublishedAnnual Conference of the International Speech Communication Association (INTERSPEECH)

  Date2023-08-21

View More

News(4)

• [Blog] Relational Proxy Loss for Audio–Text based Keyword Spotting

  Keyword spotting (KWS) aims at identifying pre-specified keywords in audio streams. Recently, KWS has attracted considerable research attention with the popularity of voice assistants that are triggered by keywords such as “Alexa,” “Hi Bixby,” or “Okay Google.” KWS techniques can be broadly classified into two categories: fixed KWS [1-3] and flexible (or user-defined) KWS [4-13].

• [Blog] Relational Proxy Loss for Audio–Text based Keyword Spotting

  Keyword spotting (KWS) aims at identifying pre-specified keywords in audio streams. Recently, KWS has attracted considerable research attention with the popularity of voice assistants that are triggered by keywords such as “Alexa,” “Hi Bixby,” or “Okay Google.” KWS techniques can be broadly classified into two categories: fixed KWS [1-3] and flexible (or user-defined) KWS [4-13].

• [Blog] Online Continual Learning in Keyword Spotting for Low-Resource Devices via Pooling High-Order Temporal Statistics

  Recent advancement in speech signal processing has produced extremely large transformer models with billions of parameters achieving outstanding performance compared to previous models.

View More

Others(0)