Audio Formats

Bitrate

Bitrate is the amount of data used per second of audio, measured in kbps — higher bitrate means better quality and larger files.

VoisLabs TeamUpdated March 2026

Bitrate is the amount of data used to encode each second of audio, measured in kilobits per second (kbps). Higher bitrate produces better audio quality and larger file sizes; lower bitrate produces smaller files with audible compression artifacts. For MP3: 64 kbps is AM radio-ish quality (sufficient for pure voice), 128 kbps is FM radio-ish (casual music), 192 kbps is near-CD-quality, 256-320 kbps approaches CD-quality. For AAC: roughly 75% of the MP3 equivalent bitrate produces similar quality — 96 kbps AAC ≈ 128 kbps MP3. For WAV: bitrate is fixed by sample rate × bit depth × channels — 44.1 kHz × 16-bit × 2 channels = 1411 kbps (10× higher than compressed formats). Voice content tolerates much lower bitrates than music because human voice frequency range is narrower — podcast and audiobook bitrates can go as low as 48-64 kbps mono without audible degradation. Variable bitrate (VBR) adjusts bitrate dynamically per audio segment, producing better overall quality at a given average bitrate than constant bitrate (CBR).

How it works

Bitrate tradeoffs: lower bitrate = smaller file + potentially audible artifacts; higher bitrate = larger file + better quality (up to a point). Beyond ~256 kbps for music and ~128 kbps for voice, bitrate increases don't produce audible improvement for most listeners — diminishing returns set in. For podcast distribution, 64-96 kbps MP3 or AAC is standard — small enough for fast download, high enough for clear voice. For music streaming, 256-320 kbps is typical. For archival or professional work, lossless (WAV or FLAC) eliminates bitrate tradeoffs by storing the full audio. Bitrate should match content type: mono voice at 64 kbps is fine; stereo music at 64 kbps sounds bad. Bitrate is independent of sample rate — you can have high-sample-rate low-bitrate (poor quality) or low-sample-rate high-bitrate (wastes bitrate) files. Most tools calculate bitrate automatically when you set a quality target; manual bitrate control is typically only needed for specific platform requirements.

Examples

Podcast at 64 kbps mono MP3

A 60-minute podcast episode = ~28 MB file. Small enough for fast download on mobile data, voice quality clear.

Music at 256 kbps AAC

Apple Music default. A 4-minute song = ~7.5 MB. Near-indistinguishable from CD quality for most listeners.

WAV fixed bitrate

CD-quality WAV (44.1 kHz, 16-bit, stereo) = 1411 kbps fixed. A 4-minute song = ~40 MB. Used for masters and archival.

Why this matters for Indian-language TTS

For Indian distribution, bitrate matters because mobile data costs are consumption-bound — smaller files reduce listener data consumption. Hindi/Tamil/Malayalam podcasts at 48-64 kbps mono are common; audiobooks often use 64 kbps for the file-size savings. VoisLabs exports MP3 at a default bitrate optimised for voice quality and file size; WAV export on Studio/Pro tier bypasses bitrate considerations entirely.

Frequently Asked Questions

What bitrate should I use for a podcast?
64-96 kbps mono MP3 or AAC for pure voice. 128 kbps stereo MP3 if you include music or multi-host sound. Going above 128 kbps for voice content produces no audible quality improvement — just larger files.
Is higher bitrate always better?
Only up to a point. Past transparent bitrate (the threshold where most listeners can't detect compression artifacts), higher bitrates just waste storage and bandwidth. For voice: transparent is around 96 kbps MP3 / 64 kbps AAC. For music: around 192 kbps MP3 / 128 kbps AAC. Beyond those, lossless or not is the only meaningful step.
What's variable vs constant bitrate?
CBR (constant bitrate) uses the same bitrate throughout — simple, predictable file size. VBR (variable bitrate) adjusts dynamically — more bits for complex passages, fewer for silence or simple tones. VBR typically produces better quality at a given average bitrate. Most modern encoders default to VBR.

Try VoisLabs — Indian-language TTS done right

1 minute free per day. 12 languages. Native Indian-script karaoke subtitles. No card required.

Start free

Last verified: 2026-04-21