Combining Media-Specific FEC and Error Concealment for Robust Distributed Speech Recognition Over Loss-Prone Packet Channels | Publicación