A voice conversion training system, voice conversion system, voice conversion client-server system, and program that realize voice conversion to be performed with low load of training are provided.In a server 10, an intermediate conversion function generation unit 101 generates an intermediate conversion function F, and a target conversion function generation unit 102 generates a target conversion function G. In a mobile terminal 20, an intermediate voice conversion unit 211 uses the conversion function F to generate speech of an intermediate speaker from speech of a source speaker, and a target voice conversion unit 212 uses the conversion function G to convert speech of the intermediate speaker speech generated by the intermediate voice conversion unit 211 to speech of a target speaker.