Method and system for automatically detecting morphemes in a task classification system using lattices
AssigneeAT&T Corp.
InventorsAllen Louis Gorin, Dijana Petrovska-Delacretaz, Giuseppe Riccardi, Jeremy Huntley Wright
Life of this patent
Estimating the term…
Abstract
In an embodiment, a lattice of phone strings in an input communication of a user may be recognized, wherein the lattice may represent a distribution over the phone strings. Morphemes in the input communication of the user may be detected using the recognized lattice. Task-type classification decisions may be made based on the detected morphemes in the input communication of the user.
Claims20
A tangible computer-readable medium storing instructions for controlling a computing device to detect morphemes using lattices in an automated task classification system which operates on one or more task objectives of a user, the instructions comprising:
- recognizing a lattice of phone strings in user input, the lattice representing a distribution over the phone strings;
- detecting morphemes in the user input using the recognized lattice;
- and
- making task-type classification decisions based on the detected morphemes in the user input.
A tangible computer-readable medium storing instructions for controlling a computing device to perform task classification which operates on one or more task objectives of a user, the instructions comprising:
- selecting salient phone-phrases from training speech;
- generating morphemes by clustering selected ones of the salient phone-phrases which are semantically and syntactically similar;
- recognizing a lattice of phone strings in user input, the lattice representing a distribution over the phone strings;
- detecting morphemes in the user input using the recognized lattice;
- and
- making task-type classification decisions based on the detected morphemes in the user input.
A tangible computer-readable medium storing instructions for controlling a computing device to perform task classification using morphemes which operate on a task objective of a user, the morphemes being generated by clustering selected ones of silent sub-morphemes or salient phone phrases from training speech which are semantically and syntactically similar, the instructions comprising:
- detecting morphemes present in user input by utilizing an input speech recognizer, the user input including verbal speech from the user;
- and
- making task-type classification decisions based on the detected morphemes in the user input, wherein the input speech recognizer detects the morphemes present in the user input.
Dependencies are inferred from the claim text.