Display timing determination device, display timing determination method, and program

A display timing determination method, comprising:

• acquiring voice storage data storing a plurality of voices to be output sequentially and a sound other than a voice;
• acquiring timing data on provisional display timings of a plurality of pieces of character information, which are to be sequentially displayed during reproduction of the voice storage data and represent content of the respective plurality of voices;
• analyzing a voice waveform of the voice storage data to acquire an output timing of each of the plurality of sounds stored in the voice storage data;
• acquiring, for each of the plurality of sounds, probability information on a probability that the each of the plurality of sounds is a voice;
• and
• determining a definitive display timing of each of the plurality of pieces of character information based on the probability information acquired for each of the plurality of sounds, the output timing of each of the plurality of voices, and the provisional display timing of the respective plurality of pieces of character information determined based on the liming data.

A display timing determination method, comprising:

• acquiring voice storage data storing a plurality of voices to be output sequentially;
• acquiring timing data on provisional display timings of a plurality of pieces of character information, which are to be sequentially displayed during reproduction of the voice storage data and represent content of the respective plurality of voices;
• analyzing a voice waveform of the voice storage data to acquire an output timing each of the plurality of voices;
• repeatedly acquiring a change amount of the provisional display timings of the plurality of pieces of character information;
• acquiring, for each change amount, overall match degree information indicating an overall degree of matching between the provisional display timings of the plurality of pieces of character information, which are obtained by being changed by the each change amount, and output timings of the respective plurality of voices;
• acquiring a new change amount when a difference between the overall match degree information having a highest degree of matching and the overall match degree information having a second highest degree of matching match degree is smaller than a threshold value;
• and
• determining, when the difference is equal to or larger than the threshold value, without acquiring the new change amount, the definitive display timing of each of the plurality of pieces of character information based on a change amount corresponding to the overall match degree information having a lowest degree of matching.

A display timing determination method, comprising:

• acquiring voice storage data storing a plurality of voices to be output sequentially;
• acquiring timing data on provisional display timings of a plurality of pieces of character information, which are to be sequentially displayed during reproduction of the voice storage data and represent content of the respective plurality of voices;
• analyzing a voice waveform of the voice storage data to acquire an output timing each of the plurality of voices;
• acquiring voice speaker information on a speaker corresponding to each of the plurality of voices;
• acquiring character-information speaker information on a speaker corresponding to each of the plurality of pieces of character information;
• determining the definitive display timing of each of the plurality of pieces of character information based additionally on the voice speaker information on each of the plurality of voices and the character-information speaker information on each of the plurality of pieces of character information.