VoxaTrace/com.musicmuni.voxatrace.sonix.model.buffer/AudioBufferPool

AudioBufferPool

class AudioBufferPool(poolSize: Int, bufferSize: Int)

Pre-allocated buffer pool for zero-allocation audio processing.

This pool provides reusable FloatArray buffers for real-time DSP operations, avoiding garbage collection pauses in the hot path.

Usage

val pool = AudioBufferPool(poolSize = 4, bufferSize = 2048)

// Acquire a buffer for processing
val buffer = pool.acquire()

// Fill with audio samples
audioBuffer.fillFloatSamples(buffer)

// Process with Calibra
calibra.detectPitch(buffer, sampleCount)

// Release back to pool
pool.release(buffer)

Thread Safety

All operations are thread-safe. Multiple threads can acquire and release buffers concurrently.

Parameters

poolSize

Number of buffers to pre-allocate

bufferSize

Size of each buffer in samples

Constructors

AudioBufferPool

constructor(poolSize: Int, bufferSize: Int)

Properties

val available: Int

Number of buffers currently available in the pool.

val totalAllocated: Int

Total buffers created (includes overflow allocations). If this exceeds poolSize, the pool was exhausted at some point.

val wasExhausted: Boolean

Check if the pool has been exhausted (more allocations than poolSize). Use this in debug builds to tune poolSize.

Functions

fun acquire(): FloatArray

Acquire a buffer from the pool.

fun release(buffer: FloatArray)

Release a buffer back to the pool.

fun resetStats()

Reset pool statistics (for debugging).