Abstract: Modern automatic speech recognition (ASR) systems can utilize artificial intelligence (AI) models to service ASR requests. The number and scale of AI models used in a modern ASR system can be substantial. The process of configuring and reconfiguring hardware to execute various AI models corresponding to a substantial number of ASR requests can be time consuming and inefficient. Among other features, the described technology utilizes batching of ASR requests, splitting of the ASR requests, and/or parallel processing to efficiently use hardware tasked with executing AI models corresponding to ASR requests. In one embodiment, the compute graphs of ASR tasks are used to batch the ASR requests. The corresponding AI models of each batch can be loaded into hardware, and batches can be processed in parallel. In some embodiments, the ASR requests are split, batched, and processed in parallel.

Abstract: An apparatus for sealing a clearance between a host pipe and a liner pipe. The liner pipe is fed into the host pipe and includes a tubular element forming a leading end of the liner pipe. It also includes a compressible seal which is adapted when compressed longitudinally to expand radially and is mounted on the tubular element when in use. The seal moves freely with the element in the direction of feeding of the liner pipe. But when the liner pipe is retracted counter to the direction of feeding, the seal movement is arrested and the seal compressed longitudinally by the element so that the seal expands radially to engage the internal wall of the host pipe and seal the clearance.