Audio identification method and apparatus, and computer storage medium

An audio identification apparatus, the apparatus comprising:

• a processor;
• a memory configured to store processor-executable instructions that, when executed by the processor, cause the processor to:
• obtain an original pitch sequence of an audio sample, the original pitch sequence being used to indicate a frequency of the audio sample at each time point;
• divide the original pitch sequence into a plurality of pitch sub-sequences by:
• determining mute sequences included in the original pitch sequence, the mute sequence being a sequence whose pitch value is 0;
• determining, according to lengths of the mute sequences, that a location of at least one of the mute sequences in the original pitch sequence is a determined division location by:
• determining whether a number of the mute sequences is greater than m, m being an integer greater than or equal to 1;
• and
• when the number of the mute sequences is not greater than m, determining that locations of the mute sequences in the original pitch sequence are division locations;
• or
• when the number of the mute sequences is greater than m, determining that locations of k mute sequences having longest lengths in the mute sequences in the original pitch sequence are division locations, k being less than or equal to m, and k being a positive integer;
• respectively identify the original pitch sequence and the plurality of pitch sub-sequences;
• and
• combine identification results obtained by respectively identifying the original pitch sequence and the plurality of pitch sub-sequences, to obtain an identification result of the audio sample.

A non-transitory computer storage medium, the non-transitory computer storage medium storing processor executable instructions that, when executed by a processor, cause the processor to:

• obtain an original pitch sequence of an audio sample, the original pitch sequence configured to indicate a frequency of the audio sample at a plurality of time points;
• divide the original pitch sequence into a plurality of pitch sub-sequences by:
• determining mute sequences included in the original pitch sequence, the mute sequence being a sequence whose pitch value is 0;
• determining, according to lengths of the mute sequences, that a location of at least one of the mute sequences in the original pitch sequence is a determined division location by:
• determining whether a number of the mute sequences is greater than m, m being an integer greater than or equal to 1;
• and
• when the number of the mute sequences is not greater than m, determining that locations of the mute sequences in the original pitch sequence are division locations;
• or
• when the number of the mute sequences is greater than m, determining that locations of k mute sequences having longest lengths in the mute sequences in the original pitch sequence are division locations, k being less than or equal to m, and k being a positive integer;
• respectively identify the original pitch sequence and the plurality of pitch sub-sequences;
• and
• combine identification results obtained by respectively identifying the original pitch sequence and the plurality of pitch sub-sequences, to obtain an identification result of the audio sample.