Abstract: Methods, systems, and apparatus, including computer programs encoded on computer-readable storage media, for speech recognition using attention-based sequence-to-sequence models. In some implementations, audio data indicating acoustic characteristics of an utterance is received. A sequence of feature vectors indicative of the acoustic characteristics of the utterance is generated. The sequence of feature vectors is processed using a speech recognition model that has been trained using a loss function that uses a set of speech recognition hypothesis samples, the speech recognition model including an encoder, an attention module, and a decoder. The encoder and decoder each include one or more recurrent neural network layers. A sequence of output vectors representing distributions over a predetermined set of linguistic units is obtained. A transcription for the utterance is obtained based on the sequence of output vectors. Data indicating the transcription of the utterance is provided.

Abstract: A network component to partition an Internet Protocol (IP) Radio Access Network (RAN) network. The network component may be configured to find one or more rings and one or more chains. The rings and/or the chains may be used to find one or more ring clusters. The network component may be configured to divide one or more of the ring clusters if they exceed a network node threshold number of network nodes. The network node may form a plurality of interior gateway protocol (IGP) areas using the ring clusters. The plurality of IGP areas may be refined based on a network node threshold and an aggregate site gateway (ASG) node threshold.