Singing voice-synthesizing method and apparatus and storage medium

• US 7,249,022 B2
• Filed: 12/01/2005
• Issued: 07/24/2007
• Est. Priority Date: 12/28/2000
• Status: Expired due to Fees

• Current Assignee: Yamaha Corporation
• Original Assignee: Yamaha Corporation
• Inventors: Kayama, Hiraku, Celma, Oscar, Ortola, Jaume
• Primary Examiner(s): Lerner; Martin
• Application Number:
  US11/292,035
• Publication Number:
  US 20060085196A1
• Time in Patent Office: 600 Days
• Field of Search: 704/258, 704/260, 704/267, 704/268, 704/269, 704/270, 846/22, 846/28, 846/59, 846/65, 846/68
• US Class Current: 704/267
• CPC Class Codes: G10H 2210/201 Vibrato, i.e. rapid, repetitive and smooth variation of amplitude, pitch or timbre within a note or chord
• 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
• CPC Class Codes: G10L 13/02 Methods for producing synthetic speech; Speech synthesisers
• CPC Class Codes: G10L 13/033 Voice editing, e.g. manipulating the voice of the synthesiser
• CPC Class Codes: G10L 13/06 Elementary speech units used in speech synthesisers; Concatenation rules

Abstract

There are provided a singing voice-synthesizing method and apparatus capable of performing synthesis of natural singing voices close to human singing voices based on performance data being input in real time. Performance data is inputted for each phonetic unit constituting a lyric, to supply phonetic unit information, singing-starting time point information, singing length information, etc. Each performance data is inputted in timing earlier than the actual singing-starting time point, and a phonetic unit transition time length is generated. By using the phonetic unit transition time, the singing-starting time point information, and the singing length information, the singing-starting time points and singing duration times of the first and second phonemes are determined. In the singing voice synthesis, for each phoneme, a singing voice is generated at the determined singing-starting time point and continues to be generated for the determined singing duration time.

Claims

1. A singing voice-synthesizing method comprising:

inputting phonetic unit information representative of a phonetic unit, time information representative of a singing-starting time point, and singing length information representative of a singing length, for a singing phonetic unit including a sequence of a first phoneme and a second phoneme;

generating a phonetic unit transition time length formed by a generation time length of the first phoneme and a generation time length of the second phoneme, based on the inputted phonetic unit information;

generating a state transition time length corresponding to a rise portion, a note transition portion, or a fall portion of the singing phonetic unit, based on the inputted phonetic unit information and the generated phonetic unit transition time length; and

generating a singing voice formed by the phonetic unit, based on the phonetic unit information, the time information, and the singing length information which have been inputted, the generating step including adding a change in at least one of pitch and amplitude to the singing voice during a time period corresponding to the generated state transition time length.

2. A singing voice-synthesizing apparatus comprising:

an input section that inputs phonetic unit information representative of a phonetic unit, time information representative of a singing-starting time point, and singing length information representative of a singing length, for a singing phonetic unit including a sequence of a first phoneme and second phoneme;

a storage section that stores state transition time length corresponding to a rise portion, a note transition portion, or a fall portion of the singing phonetic unit, the state transition time length being generated based on inputted phonetic unit information and phonetic unit transition time length formed by a generation time length of the first phoneme and a generation time length of the second phoneme, based on the inputted phonetic unit information;

a readout section that reads out the state transition time length from said storage section based on the phonetic unit information inputted by said input section; and

a singing voice-synthesizing section that generates a singing voice formed by the phonetic unit, based on the phonetic unit information, the time information, and the singing length information which have been inputted by said input section, said singing voice-synthesizing section adding a change in at least one of pitch and amplitude to the singing voice during a time period corresponding to the state transition time length read out by said readout section.

3. A singing voice-synthesizing apparatus according to claim 2, wherein said input section inputs modifying information for modifying the state transition time length, and wherein the singing voice-synthesizing apparatus includes a modifying section that modifies the state transition time length read out by said readout section based on the modifying information inputted by said input section, and wherein said singing voice-synthesizing section adds a change in at least one of pitch and amplitude to the singing voice during a time period corresponding to the state transition time length modified by said modifying section.

4. A storage medium storing a program for executing a singing voice-synthesizing method, the program comprising:

an input module that inputs phonetic unit information representative of a phonetic unit, time information representative of a singing-starting time point, and singing length information representative of a singing length, for a singing phonetic unit including a sequence of a first phoneme and a second phoneme;

a phonetic unit transition time length generating module that generates a phonetic unit transition time length formed by a generation time length of the first phoneme and a generation time length of the second phoneme, based on the inputted phonetic unit information

a state transition time length-generating module that generates a state transition time length corresponding to a rise portion, a note transition portion, or a fall portion of the singing phonetic unit, based on the inputted phonetic unit information and the generated phonetic unit transition time length; and

a singing voice-generating module that generates a singing voice formed by the phonetic unit, based on the phonetic unit information, the time information, and the singing length information which have been inputted, the singing voice-generating module adding a change in at least one of pitch and amplitude to the singing voice during a time period corresponding to the generated state transition time length.