Character curve generating method and device thereof

A character curve generating method, the character curve referring to a text formed by characters of different sizes, wherein the method comprises the following steps:

• (1) inputting text information, wherein the text information comprises m characters, and m is a positive integer;
• (2) performing voice input to generate an audio file;
• (3) dividing the audio file into m equal time periods and each character corresponds to multiple sample values, all the sample values are converted into PCM (Pulse Code Modulation) codes, absolute values of this set of PCM codes are compared, to find a maximum PCM value, and the maximum PCM value is stored as a sound volume of the time period;
• (4) calculating dimension information corresponding to each character in the text information according to sound volumes of different time periods in the audio file;
• wherein the calculating dimension information comprises the following steps:
• (41) reading q font sizes forming the character curve, wherein q is a positive integer;
• (42) setting a minimum PCM value to Pmin and the maximum PCM value to Pmax;
• and
• (43) calculating PCM values Pn respectively corresponding to the q font sizes:
• Pn=Pmin+(n−1)*d, wherein
• d=(Pmax−Pmin)(q−1), and n=1, . . . , and q;
• and
• (44) calculating a font size according to a PCM value corresponding to each character, wherein
• when the corresponding PCM value is between Pn and Pn-1, if the PCM value≥(Pn−Pn-1)/2, a font size corresponding to Pn is used for the character;
• or if the PCM value<(Pn−Pn-1)/2, a font size corresponding to Pn-1 is used for the character;
• and
• (5) changing display sizes of the m characters in the text information according to the dimension information acquired in step (4) to form the character curve.