Information processing apparatus, output control method, and computer-readable recording medium

An information processing apparatus comprising:

• a processor configured to:
• receive a phoneme signal for a text having a plurality of sentences;
• performing Fourier transformation on the received phoneme signal to perform spectral analysis and extracting voice features;
• estimating a phoneme string from the received phoneme signal based on the extracted voice features and registering a unit separator indicating separation between estimated phoneme strings;
• determine a sentence corresponding to a specific phoneme or the phoneme string included in the received phoneme signal;
• refer to a storage that stores therein co-occurrence information on sentences for words in association with the words and determine a word the co-occurrence information on the determined sentence of which satisfies a standard among the words, the co-occurrence information on sentences being information including vector information corresponding to the sentence, the vector information including a word vector that is calculated from an encoded word based on a Word2Vec technique and a sentence vector that is calculated by accumulating the word vectors corresponding to words included in a sentence corresponding to the sentence vector;
• change the specific phoneme or the phoneme string included in the received phoneme signal to the determined word to generate a text corresponding to the received phoneme signal;
• and
• output the text generated at the generating.

A output control method comprising:

• receiving a phoneme signal for a text having a plurality of sentences;
• performing Fourier transformation on the received phoneme signal to perform spectral analysis and extracting voice features;
• estimating a phoneme string from the received phoneme signal based on the extracted voice features and registering a unit separator indicating separation between estimated phoneme strings;
• determining a sentence corresponding to a specific phoneme or the phoneme string included in the received phoneme signal;
• referring to a storage that stores therein co-occurrence information on sentences for words in association with the words and determining a word the co-occurrence information on the determined sentence of which satisfies a standard among the words, the co-occurrence information on sentences being information including vector information corresponding to the sentence, the vector information including a word vector that is calculated from an encoded word based on a Word2Vec technique and a sentence vector that is calculated by accumulating the word vectors corresponding to words included in a sentence corresponding to the sentence vector;
• changing the specific phoneme or the phoneme string included in the received phoneme signal to the determined word to generate a text corresponding to the received phoneme signal, by a processor;
• and
• outputting the generated text.