Query-by-Example Spoken Term Detection (QbE STD) aims at retrieving data from a speech data repository given an\r\nacoustic query containing the term of interest as input. Nowadays, it has been receiving much interest due to the high\r\nvolume of information stored in audio or audiovisual format. QbE STD differs from automatic speech recognition (ASR)\r\nand keyword spotting (KWS)/spoken term detection (STD) since ASR is interested in all the terms/words that appear in\r\nthe speech signal and KWS/STD relies on a textual transcription of the search term to retrieve the speech data. This\r\npaper presents the systems submitted to the ALBAYZIN 2012 QbE STD evaluation held as a part of ALBAYZIN 2012\r\nevaluation campaign within the context of the IberSPEECH 2012 Conferencea. The evaluation consists of retrieving\r\nthe speech files that contain the input queries, indicating their start and end timestamps within the appropriate\r\nspeech file. Evaluation is conducted on a Spanish spontaneous speech database containing a set of talks from MAVIR\r\nworkshopsb, which amount at about 7 h of speech in total. We present the database metric systems submitted along\r\nwith all results and some discussion. Four different research groups took part in the evaluation. Evaluation results\r\nshow the difficulty of this task and the limited performance indicates there is still a lot of room for improvement. The\r\nbest result is achieved by a dynamic time warping-based search over Gaussian posteriorgrams/posterior phoneme\r\nprobabilities. This paper also compares the systems aiming at establishing the best technique dealing with that\r\ndifficult task and looking for defining promising directions for this relatively novel task.
Loading....