Skip to main content
EURASIP Journal on Audio, Speech, and Music Processing

Table 1 Search-on-speech applications

From: ALBAYZIN 2016 spoken term detection evaluation: an international open competitive evaluation in Spanish

Type of result (output)

Type of query (input)

Text

Query

Document

SDR

QbE-SDR

Document+Position

STD/KWS

QbE-STD

  1. ‘SDR’ stands for Spoken Document Retrieval, ‘STD’ for Spoken Term Detection, ‘KWS’ for Keyword Spotting, and ‘QbE’ for Query-by-Example. STD is called when the audio is processed before knowing the terms to search, and KWS knows beforehand the terms to search
Back to article page