Speech synthesizing device, speech synthesizing method, and program

• US 8,209,180 B2
• Filed: 02/01/2007
• Issued: 06/26/2012
• Est. Priority Date: 02/08/2006
• Status: Active Grant

• Current Assignee: NEC Corporation
• Original Assignee: NEC Corporation
• Inventors: Kato, Masanori
• Primary Examiner(s): Albertalli, Brian
• Application Number:
  US12/223,707
• Publication Number:
  US 20100145706A1
• Time in Patent Office: 1,972 Days
• Field of Search: None
• US Class Current: 704/258
• CPC Class Codes: G10H 2240/081 Genre classification, i.e. descriptive metadata for classification or selection of musical pieces according to style analysis genre G10H2210/036
• CPC Class Codes: G10H 2250/455 Gensound singing voices, i.e. generation of human voices for musical applications, vocal singing sounds or intelligible words at a desired pitch or with desired vocal effects, e.g. by phoneme synthesis formant synthesis G10H2250/481; parcor synthesis G10H2250/505; modulation effects G10H2210/195; ensemble effects G10H2210/245; speech synthesis in general G10L13/00
• CPC Class Codes: G10L 13/10 Prosody rules derived from text; Stress or intonation

Abstract

An object of the present invention is to provide a device and a method for generating a synthesized speech that has an utterance form that matches music. A musical genre estimation unit of the speech synthesizing device estimates the musical genre to which a received music signal belongs, an utterance form selection unit references an utterance form information storage unit to determine an utterance form from the musical genre. A prosody generation unit references a prosody generation rule storage unit, selected from prosody generation rule storage units 15₁ to 15_N according to the utterance form, and generates prosody information from a phonetic symbol sequence. A unit waveform selection unit references a unit waveform data storage unit, selected from unit waveform data storage units 16₁ to 16_N according to the utterance form, and selects a unit waveform from the phonetic symbol sequence and the prosody information. A waveform generation unit generates a synthesized speech waveform from the prosody information and the unit waveform data.

Claims

1. A speech synthesizing device comprising:

an utterance form selection unit that analyzes a music signal reproduced in a user environment and determines an utterance form that matches an analysis result of the music signal;

a speech synthesizing unit that synthesizes a speech according to the utterance form;

a music signal power calculation unit that analyzes the music signal and calculates a power of the music signal;

a synthesized speech power calculation unit that analyzes the synthesized speech waveform and calculates a power of the synthesized speech; and

a synthesized speech power adjustment unit that references a ratio predetermined for each utterance form between a power of the music signal and a power of the synthesized speech and adjusts a power of the synthesized speech waveform, generated according to the utterance form, according to the power of the music signal.

2. A speech synthesizing method that generates a synthesized speech using a speech synthesizing device, said method comprising:

analyzing, by said speech synthesizing device, a music signal reproduced in a user environment and determining an utterance form that matches an analysis result of the music signal;

synthesizing, by said speech synthesizing device, a speech according to the utterance form;

analyzing, by said speech synthesizing device, the music signal and calculating a power of the music signal;

analyzing, by said speech synthesizing device, the synthesized speech waveform and calculating a power of the synthesized speech; and

referencing, by said speech synthesizing device, a ratio predetermined for each utterance form between a power of the music signal and a power of the synthesized speech and adjusting a power of the synthesized speech waveform, generated according to the utterance form, according to the power of the music signal.

3. A non-transitory computer readable medium storing a computer program causing a computer, which constitutes a speech synthesizing device, to execute:

processing for analyzing a received music signal reproduced in a user environment and determining an utterance form, which matches an analysis result of the music signal, from utterance forms prepared in advance;

processing for synthesizing a speech according to the utterance form;

processing for analyzing the music signal and estimating a musical genre to which the music belongs;

processing for selecting an utterance form according to the musical genre to determine the utterance form that matches the analysis result of the music signal;

processing for analyzing the music signal and calculating a power of the music signal;

processing for analyzing the synthesized speech waveform and calculating a power of the synthesized speech; and

processing for referencing a ratio predetermined for each utterance form between a power of the music signal and a power of the synthesized speech and adjusting a power of the synthesized speech waveform, generated according to the utterance form, according to the power of the music signal.