Parental state and offspring recognition in the biparental cichlid fish Pelvicachromis pulcher

Alloparental care was investigated in the biparental West African cichlid, Pelvicachromis pulcher. Non-breeding adults typically consumed young conspecifics but this trait was inhibited in both sexes during reproductive attempts. Alien conspecific young were accepted into the brood if they were of a similar age/developmental stage to the parents' own young but not if they were much older or much younger. If not accepted they were consumed by whichever adult located them. Parents separated from their brood for up to 4 days accepted their young on reunion but separation for more than 4 days resulted in the young being consumed. This latter response occurred if chemical stimuli from the young were available during the separation but not if visual stimuli were available. In this latter case parental responsiveness was maintained. Both sexes of this externally fertilizing species appeared to have the same information about their young and showed the same changes in responsiveness and the same discriminatory abilities. (C) 1997 The Association for the Study of Animal Behaviour.

Speaker Recognition In Noisy Conditions With Limited Training Data

In this paper we present a novel method for performing speaker recognition with very limited training data and in the presence of background noise. Similarity-based speaker recognition is considered so that speaker models can be created with limited training speech data. The proposed similarity is a form of cosine similarity used as a distance measure between speech feature vectors. Each speech frame is modelled using subband features, and into this framework, multicondition training and optimal feature selection are introduced, making the system capable of performing speaker recognition in the presence of realistic, time-varying noise, which is unknown during training. Speaker identi?cation experiments were carried out using the SPIDRE database. The performance of the proposed new system for noise compensation is compared to that of an oracle model; the speaker identi?cation accuracy for clean speech by the new system trained with limited training data is compared to that of a GMM trained with several minutes of speech. Both comparisons have demonstrated the effectiveness of the new model. Finally, experiments were carried out to test the new model for speaker identi?cation given limited training data and with differing levels and types of realistic background noise. The results have demonstrated the robustness of the new system.