Uncompressed PCM audio size calculator

Frames make
the file.

Start with sample rate, stored sample width, channels and duration. Get exact payload bytes, clearly labelled decimal and binary sizes, and a local plan that keeps container overhead separate.

Exact integer byte countsNo recording or uploadFree, browser-local

Plan one PCM payload

ENTER FORMAT · COVER COMPLETE FRAMES · KEEP BYTES EXACT

01 / Enter the stored format

Count what is stored.

This is an uncompressed, byte-aligned PCM model. Enter the stored width, which may exceed the valid signal precision. Suggested sample rates are arithmetic inputs, not a device-support claim.

Sample rate per channel HzWhole samples per second for each channel, from 1 to 10,000,000.

Stored bits per channel sampleChannels1–4,096. This does not assign speaker positions.

Valid signal bits optionalIf known, enter 1 up to the stored width. This changes the explanation, never the payload bytes. For 24 valid bits in a 32-bit container, select 32 stored bits.Requested durationNonnegative plain decimal, up to 12 decimal places. Enough complete frames are rounded up to cover this duration. This is our planning convention, not every recorder's rule.

Additional non-payload bytes optional · BEnter known header, metadata, padding or other non-payload bytes for this one file. Blank means combined size unavailable; 0 means known zero. No universal 44-byte header is assumed.

02 / Exact stored payload

Uncompressed audio data

BYTES · EXCLUDES NON-PAYLOAD DATA

Complete frames
Bytes per frame
Payload byte rate
Payload bit rate

Covered duration:

Different unit views of the exact payload byte count
Unit	Payload size	Display

Keep the container separate

Additional non-payload bytes
Combined nominal bytes

Byte counts use exact integers. Rounded unit views never change bytes or the scalar-field check. This is a size model; no audio was analysed.

One uncompressed byte-aligned PCM payload. No compressed-codec prediction, recording, upload, filesystem allocation, backup-copy or device-compatibility calculation. The 32-bit check concerns one scalar field only, not complete WAV fit. Arithmetic limits: duration up to 1,000,000,000,000 seconds and optional additional bytes up to 10³⁰ B; equivalent duration limits apply in all units.

Use stored sample width.

Valid precision is explanatory. A 24-bit signal inside a 32-bit container still occupies four bytes per channel sample in this model.

Read the unit labels.

MB uses 1,000,000 bytes. MiB uses 1,048,576 bytes. Both views describe the same exact payload, with display rounding labelled.

Keep unknowns visible.

Blank additional bytes leave the combined file size unavailable. Entering zero is a different, explicit assumption. Exports preserve that distinction.