Go Split WAV File

split.go

package main

import (
	"fmt"
	"github.com/go-audio/wav"
	"github.com/lunabrain-ai/lunabrain/pkg/store/bucket"
	"github.com/pkg/errors"
	"math"
	"os"
	"path"
)

func splitWAVFile(fileStore *bucket.Bucket, id string, filepath string, maxFileSize int, cb func(string) error) error {
	headerSize := 44 + 1024
	file, err := os.Open(filepath)
	if err != nil {
		return err
	}
	defer file.Close()

	decoder := wav.NewDecoder(file)

	buf, err := decoder.FullPCMBuffer()
	if err != nil {
		return err
	}

	frameCount := buf.NumFrames()
	sampleSize := (buf.SourceBitDepth / 8) * buf.Format.NumChannels
	maxSamplesPerChunk := (maxFileSize - headerSize) / sampleSize

	chunks := int(math.Ceil(float64(frameCount) / float64(maxSamplesPerChunk)))

	d, err := fileStore.NewDir(id + "_chunks")
	if err != nil {
		return errors.Wrapf(err, "failed to create bucket dir for file %s", filepath)
	}

	for i := 0; i < chunks; i++ {
		start := i * maxSamplesPerChunk
		end := start + maxSamplesPerChunk
		if end > frameCount {
			end = frameCount
		}

		bucketFile := path.Join(d, fmt.Sprintf("chunk_%d.wav", i))
		outFile, err := os.Create(bucketFile)
		if err != nil {
			return errors.Wrapf(err, "failed to create file")
		}

		enc := wav.NewEncoder(outFile, buf.Format.SampleRate, buf.SourceBitDepth, buf.Format.NumChannels, 1)

		// take a slice of the audio buffer for the chunk
		b := buf.Clone().AsIntBuffer()
		b.Data = buf.Data[start*buf.Format.NumChannels : end*buf.Format.NumChannels]

		err = enc.Write(b)
		if err != nil {
			return err
		}

		if err := enc.Close(); err != nil {
			return err
		}

		err = cb(bucketFile)
		if err != nil {
			return err
		}
	}
	return nil
}

func main() {
  fileStore, err := bucket.New(bucket.Config{
		Path: "/tmp/audio",
	})
	if err != nil {
    panic(err)
  }
	err = splitWAVFile(fileStore, "test", "test.wav", 25*1024*1024, func(chunkFilePath string) error {
    println("chunk", chunkFilePath)
	})
	if err != nil {
    panic(err)
	}
}