Multi-mode input method editor

A computer-implemented method, comprising:

• receiving, at a computing device having one or more processors, composition inputs;
• determining, at the computing device, a set of candidate selections based on the composition inputs, each candidate selection of the set of candidate selections corresponding to a representation of the composition inputs in one language of a plurality of languages;
• determining, at the computing device, a first language context value for each candidate selection, each candidate selection being associated with a specific language of the plurality of languages, each first language context value indicating a likelihood that its associated candidate selection corresponds to the composition inputs when compared against other candidate selections in the specific language;
• determining, at the computing device, a second language context value for each of the plurality of languages based on the composition inputs, each second language context value indicating a likelihood that the composition inputs correspond to its associated language of the plurality of languages;
• and
• providing for display, from the computing device, a list of candidate selections based on at least one of:
• (i) the first language context values, and (ii) the second language context values, the list of candidate selections including at least one candidate selection in each of two or more of the plurality of languages,
• wherein the list of candidate selections is a single, interleaved list of candidate selections that (i) includes at least one candidate selection in each of the two or more of the plurality of languages and (ii) identifies a rank for each of the candidate selections, each rank being indicative of a relative likelihood that its corresponding candidate selection was intended from the composition inputs.

A computing device having one or more processors and configured to execute operations comprising:

• receiving composition inputs;
• determining a set of candidate selections based on the composition inputs, each candidate selection of the set of candidate selections corresponding to a representation of the composition inputs in one language of a plurality of languages;
• determining a first language context value for each candidate selection, each candidate selection being associated with a specific language of the plurality of languages, each first language context value indicating a likelihood that its associated candidate selection corresponds to the composition inputs when compared against other candidate selections in the specific language;
• determining a second language context value for each of the plurality of languages based on the composition inputs, each second language context value indicating a likelihood that the composition inputs correspond to its associated language of the plurality of languages;
• and
• providing for display a list of candidate selections based on at least one of:
• (i) the first language context values, and (ii) the second language context values, the list of candidate selections including at least one candidate selection in each of two or more of the plurality of languages,
• wherein the list of candidate selections is a single, interleaved list of candidate selections that (i) includes at least one candidate selection in each of the two or more of the plurality of languages and (ii) identifies a rank for each of the candidate selections, each rank being indicative of a relative likelihood that its corresponding candidate selection was intended from the composition inputs.