# US 2016/0365086 A1 · Information providing device and information providing method

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2016/0365086 A1
- Document ID: US20160365086A1
- Status: Published application
- Kind: A1
- Application number: 15/120,966
- Filed: 2014-04-15
- Published: 2016-12-15
- Earliest claimed: 2014-04-15
- Assignees: Mitsubishi Electric Corporation
- Applicants: Mitsubishi Electric Corporation
- Inventors: Masanobu OSAWA; Yohei OKATO
- Web page: https://patentgrep.com/patent/US20160365086A1

## CPC

- G10L 13/08: Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
- G06F 40/205: Parsing
- G06F 40/247: Thesauruses; Synonyms
- G06F 40/279: Recognition of textual entities
- G06F 40/30: Semantic analysis
- G06F 40/253: Grammatical analysis; Style critique
- G10L 13/00: Speech synthesis; Text to speech systems
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A detector 5 detects words acoustically similar to each other from text information, and a selector 7 selects a synonym from a storage 6, the synonym corresponding to a word detected by the detector 5 and being acoustically neither the same as nor similar to any word existing in the text information. A replacer 8 replaces the word detected by the detector 5 with the synonym selected by the selector 7.

## Claims

### Claim 1

An information providing device that provides text information for voice synthesis for a voice synthesis device, said information providing device comprising:
an acquirer to acquire text information;
a detector to detect words acoustically similar to each other from said text information;
a storage to store a synonym which is brought into correspondence with each of words in advance;
a selector to select a synonym from said storage, said synonym corresponding to a word detected by said detector and being acoustically neither same as nor similar to any word existing in said text information;
a replacer to replace the word detected by said detector with the synonym selected by said selector; and
an outputter to output said text information after being replaced by said replacer as said text information for voice synthesis.

### Claim 2 (depends on 1)

The information providing device according to claim 1, wherein said information providing device includes a partitioner to partition the text information acquired by said acquirer into two or more units such as sentences, said detector detects words acoustically similar to each other from the units after being partitioned by said partitioner, and said selector selects a synonym from said storage, said synonym corresponding to a word detected by said detector and being acoustically neither same as nor similar to any word existing in the units after being partitioned by said partitioner.

### Claim 3 (depends on 2)

The information providing device according to claim 2, wherein said partitioner partitions said text information on a basis of additional information added to said text information.

### Claim 4 (depends on 1)

The information providing device according to claim 1, wherein when a plurality of synonym candidates exist, said selector selects a synonym having a lower number of phonemes.

### Claim 5

An information service method which an information providing device uses to provide a voice synthesis device with text information for voice synthesis, said information service method comprising the steps of:
in an acquirer, acquiring text information;
in a detector, detecting words acoustically similar to each other from said text information;
in a selector, selecting a synonym from synonyms which are respectively brought into correspondence with words in advance, said synonym corresponding to a word detected in said detecting step and being acoustically neither same as nor similar to any word existing in said text information;
in a replacer, replacing the word detected in said detecting step with the synonym selected in said selecting step; and
in an outputter, outputting said text information after being replaced in said replacing step as said text information for voice synthesis.
