Unified framework for text conversion and prediction

A non-transitory computer-readable storage medium storing one or more programs configured to be executed by one or more processors of an electronic device with a display and memory, the one or more programs including instructions for:

• receiving a current character input of a character input sequence, the character input sequence representing characters of a first writing system;
• determining a first current character context in the first writing system based on the received current character input and a first previous character context in the first writing system;
• determining a second current character context in a second writing system based on the first current character context in the first writing system, a second previous character context in the second writing system, and a character representation in the second writing system for the current character input;
• determining a current word context in the second writing system based on the second current character context in the second writing system, a previous word context in the second writing system, and a word representation in the second writing system for the character representation;
• determining, based on the current word context in the second writing system, a probability distribution over a word inventory in the second writing system;
• determining, based on the probability distribution over the word inventory, one or more word sequences in the second writing system for the character input sequence;
• and
• providing the one or more word sequences for output.

A method comprising:

• at an electronic device having one or more processors and memory:
• receiving a current character input of a character input sequence, the character input sequence representing characters of a first writing system;
• determining a first current character context in the first writing system based on the received current character input and a first previous character context in the first writing system;
• determining a second current character context in a second writing system based on the first current character context in the first writing system, a second previous character context in the second writing system, and a character representation in the second writing system for the current character input;
• determining a current word context in the second writing system based on the second current character context in the second writing system, a previous word context in the second writing system, and a word representation in the second writing system for the character representation;
• determining, based on the current word context in the second writing system, a probability distribution over a word inventory in the second writing system;
• determining, based on the probability distribution over the word inventory, one or more word sequences in the second writing system for the character input sequence;
• and
• providing the one or more word sequences for output.