Aiming to minimize service delay, we propose a new random caching scheme in device-to-device (D2D)-assisted heterogeneous network. To support diversified viewing qualities of multimedia video services, each video file is encoded into a base layer (BL) and multiple enhancement layers (ELs) by scalable video coding (SVC). A super layer, including the BL and several ELs, is transmitted to every user. We define and quantify the service delay of multi-quality videos by deriving successful transmission probabilities when a user is served by a D2D helper, a small-cell base station (SBS) and a macro-cell base station (MBS). We formulate a delay minimization problem subject to the limited cache sizes of D2D helpers and SBSs. The structure of the optimal solutions to the problem is revealed, and then an improved standard gradient projection method is designed to effectively obtain the solutions. Both theoretical analysis and Monte-Carlo simulations validate the successful transmission probabilities. Compared with three benchmark caching policies, the proposed SVC-based random caching scheme is superior in terms of reducing the service delay.
We present Deep Speaker, a neural speaker embedding system that maps utterances to a hypersphere where speaker similarity is measured by cosine similarity. The embeddings generated by Deep Speaker can be used for many tasks, including speaker identification, verification, and clustering. We experiment with ResCNN and GRU architectures to extract the acoustic features, then mean pool to produce utterance-level speaker embeddings, and train using triplet loss based on cosine similarity. Experiments on three distinct datasets suggest that Deep Speaker outperforms a DNN-based i-vector baseline. For example, Deep Speaker reduces the verification equal error rate by 50% (relatively) and improves the identification accuracy by 60% (relatively) on a text-independent dataset. We also present results that suggest adapting from a model trained with Mandarin can improve accuracy for English speaker recognition.
Speech data is crucially important for speech recognition research. There are quite some speech databases that can be purchased at prices that are reasonable for most research institutes. However, for young people who just start research activities or those who just gain initial interest in this direction, the cost for data is still an annoying barrier. We support the `free data' movement in speech recognition: research institutes (particularly supported by public funds) publish their data freely so that new researchers can obtain sufficient data to kick of their career. In this paper, we follow this trend and release a free Chinese speech database THCHS-30 that can be used to build a full- edged Chinese speech recognition system. We report the baseline system established with this database, including the performance under highly noisy conditions.