US 12,394,406 B2
Paralinguistic information estimation model learning apparatus, paralinguistic information estimation apparatus, and program
Atsushi Ando, Tokyo (JP); Hosana Kamiyama, Tokyo (JP); and Satoshi Kobashikawa, Tokyo (JP)
Assigned to NIPPON TELEGRAPH AND TELEPHONE CORPORATION, Tokyo (JP)
Appl. No. 17/428,961
Filed by NIPPON TELEGRAPH AND TELEPHONE CORPORATION, Tokyo (JP)
PCT Filed Jan. 27, 2020, PCT No. PCT/JP2020/002649
§ 371(c)(1), (2) Date Aug. 5, 2021,
PCT Pub. No. WO2020/162239, PCT Pub. Date Aug. 13, 2020.
Claims priority of application No. 2019-021332 (JP), filed on Feb. 8, 2019.
Prior Publication US 2022/0122584 A1, Apr. 21, 2022
Int. Cl. G10L 25/63 (2013.01); G10L 15/02 (2006.01); G10L 15/06 (2013.01); G10L 15/16 (2006.01); G10L 15/07 (2013.01); G10L 25/03 (2013.01); G10L 25/06 (2013.01); G10L 25/09 (2013.01); G10L 25/12 (2013.01); G10L 25/18 (2013.01); G10L 25/21 (2013.01); G10L 25/24 (2013.01); G10L 25/30 (2013.01)
CPC G10L 15/06 (2013.01) [G10L 15/02 (2013.01); G10L 15/16 (2013.01); G10L 25/63 (2013.01); G10L 15/063 (2013.01); G10L 15/075 (2013.01); G10L 25/03 (2013.01); G10L 25/06 (2013.01); G10L 25/09 (2013.01); G10L 25/12 (2013.01); G10L 25/18 (2013.01); G10L 25/21 (2013.01); G10L 25/24 (2013.01); G10L 25/30 (2013.01)] 12 Claims
OG exemplary drawing
 
1. A paralinguistic information estimation model learning device comprising:
an anti-teacher determiner configured to determine, based on a paralinguistic information label indicating a determination result of paralinguistic information given by a plurality of listeners for each utterance, an anti-teacher label indicating an anti-teacher serving as incorrect paralinguistic information for the utterance;
an anti-teacher estimation model learner configured to learn, based on an acoustic feature extracted from the utterance and the anti-teacher label, an anti-teacher estimation model,
wherein the anti-teacher estimation model, subsequent to the learning, estimates a posterior probability of anti-teacher from an acoustic feature of an input utterance to further estimate paralinguistic information of the input utterance based on the posterior probability.