摘要:
A method for providing variable root note support in an audio player is described. A file with Musical Instrument Digital Interface (MIDI) data and a set of user defined instruments is received. A metric is determined using a user defined root note in the user defined instruments, a key number for a MIDI note in the MIDI data, and a player specific root note. The key number is adjusted based on the metric.
摘要:
A method for providing multi-region instrument support in an audio player is described. A set of user defined instruments and a set of MIDI data is received. At least one multi-region instrument in the set of user defined instruments is identified. Each region in the multi-region instruments is mapped to a single region instrument with the same parameters as the region. A new instrument number is assigned to each single region instrument. The MIDI data is modified based on the mapping and assigning.
摘要:
A method for improving the similarity of the volumes in different audio players is described. First player metrics for one or more Musical Instrument Digital Interface (MIDI) instruments may be determined. A digital music file that uses the MIDI protocol may be received. A note parameter or channel parameter may be adjusted for notes in the digital music file based on the first player metrics.
摘要:
This disclosure describes techniques for processing audio files that comply with the musical instrument digital interface (MIDI) format. In particular, this disclosure describes storage of MIDI parameters for efficient access by a processor and a hardware unit. The processor may be a digital signal processor (DSP) and the hardware unit may be specifically designed to process MIDI parameters. In one aspect, this disclosure provides an apparatus comprising a processor that converts a MIDI event into MIDI parameters, a hardware unit that uses MIDI parameters to generate audio samples, and a plurality of storage units that store MIDI parameters which are accessible by both the processor and the hardware unit.
摘要:
This disclosure describes techniques that make use of a summing buffer that receives waveform samples from audio processing elements, and sums and stores the waveform sums for a given frame. In one example, a method comprises summing a waveform sample received from an audio processing element to produce a waveform sum associated with a first audio frame, storing the waveform sum in a memory, wherein the memory is logically partitioned into a plurality of memory blocks, and locking memory blocks containing the waveform sum associated with the first audio frame, transferring contents of locked memory blocks to an external processor, unlocking a memory block after contents of the memory block have been transferred to the external processor, and storing a waveform sum associated with a second audio frame within the unlocked memory block concurrently with transferring contents of remaining locked memory blocks associated with the first audio frame.
摘要:
Techniques are described of efficiently identifying sets of audio parameters to be applied during a time frame. For example, a list of indicators may be generated. Each of the indicators in the list may indicate a Musical Instrument Digital Interface (MIDI) voice present in a MIDI frame. Furthermore, in generating the list, the indicators in the list may be restricted to those indicators that indicate the most acoustically significant MIDI voices in the MIDI frame. After the list is generated, a digital waveform may be generated for each of MIDI voices indicated by an indicator in the list. A combination of the waveforms of each MIDI voice may constitute an overall waveform for the MIDI frame.