WAV vs MP3 vs M4A vs OGG: Which Audio Format Should You Use?

Why Audio Formats Actually Matter

Most people do not think about audio formats until something goes wrong. The podcast guest sends you a file and your editing software cannot open it. You upload a track to a streaming platform and the quality sounds oddly flat. You try to play a file on your car stereo and get silence. You download what you think is high-quality audio and it sounds worse than the YouTube video you took it from.

These are all format problems. And they are entirely preventable once you understand what each format actually does.

The stakes are different depending on what you are doing with audio. For a casual listener streaming music from Spotify, the format is completely invisible. Spotify makes the decision for you. But the moment you start recording, editing, archiving, sharing, uploading to specific platforms, or downloading audio from YouTube, format choices have real consequences.

Consequences like: your audio file is 10x larger than it needs to be, wasting storage and upload time. Or your audio is permanently degraded because you re-encoded a lossy file multiple times during editing. Or your carefully recorded interview sounds great on your laptop but thin and artifact-y when played through decent speakers because you exported at too low a bitrate.

Understanding the formats takes about 20 minutes. That knowledge saves you from hours of frustration and irreversible quality mistakes. Worth the read.

The Key Concept: Lossless vs. Lossy

Everything in audio format land starts here. There are two fundamental approaches to storing audio data.

Lossless means every single bit of the original audio data is preserved. When you decode a lossless audio file, you get back an exact bitwise copy of the original PCM (Pulse Code Modulation) audio. Play it a million times, copy it a million times, the data does not change. WAV and FLAC are lossless formats. ALAC (Apple Lossless) is another one.

Lossy means the encoder permanently discards some audio data to reduce the file size. This data is gone forever. You cannot recover it. The question is how much data, and which data, is discarded, because that determines whether the loss is audible. MP3, M4A/AAC, OGG Vorbis, and OGG Opus are all lossy formats.

The clever thing about lossy compression is that it uses psychoacoustic models, mathematical descriptions of how human hearing works, to decide which audio data to discard. Human ears are not perfect recorders. We are bad at hearing very quiet sounds that occur immediately after very loud sounds (temporal masking). We are bad at hearing certain frequencies when other nearby frequencies are very loud (frequency masking). Lossy codecs deliberately discard the audio information we are least likely to notice.

At high enough bitrates, a good lossy codec is perceptually transparent. Meaning the lossy file sounds identical to the lossless original to virtually every human listener in double-blind tests. The file is much smaller. The quality loss is real but inaudible. At lower bitrates, the loss becomes audible, ranging from subtle high-frequency softness to gargling artifacts and pre-echo distortion at very low bitrates.

The practical implication: lossless is not automatically better for listening. A 320 kbps MP3 and a WAV of the same content will sound identical to 99% of listeners on 99% of playback systems. What lossless gives you is future-proofing, re-encoding flexibility, and the assurance that nothing has been discarded.

WAV Explained

WAV (Waveform Audio File Format) was developed by Microsoft and IBM in 1991 and it is, in audio terms, extremely boring. That is a compliment. WAV is basically raw PCM audio data with a small header describing the sample rate, bit depth, and channel count. There is almost no processing, no compression, no tricks. What you recorded is what you get.

Standard CD-quality WAV is 16-bit, 44,100 Hz (44.1 kHz), stereo. That means 44,100 samples per second, each sample stored as a 16-bit number, times 2 channels. The math: 44,100 samples/sec x 16 bits x 2 channels = 1,411,200 bits per second, or about 1,411 kbps. A three-minute stereo song at CD quality is approximately 31 MB as a WAV file.

Professional audio production commonly uses 24-bit at 48 kHz or 96 kHz. Video production typically targets 24-bit 48 kHz for the audio track. At 24-bit 48 kHz stereo, a three-minute file is about 50 MB. At 32-bit float (common in DAW software for internal processing), it is even larger.

WAV has essentially no metadata support. You cannot reliably embed album art, track title, or artist information in a WAV file in a way that every player will read correctly. ID3 tags are technically possible in WAV but support is inconsistent. This is a genuine limitation if you are distributing music files.

WAV is universally compatible. Every piece of audio software ever made reads WAV. It is the common denominator. If someone asks for an audio file and you are not sure what format they need, sending WAV is almost always safe.

When to use WAV: during recording sessions (your DAW should record to WAV or AIFF), as the master file before mixing and mastering, when handing off audio to a video editor or post-production facility, and when a client or platform specifically requests uncompressed audio. Not for everyday storage or distribution. The file sizes are simply too large for that.

MP3 Explained

MP3 (MPEG-1 Audio Layer III) was standardized in 1993 and for the next 20 years was synonymous with digital audio. If you were alive and listening to music in the 2000s, you used MP3s. Your brain associates MP3 with music like it associates JPEG with photos.

MP3 uses a psychoacoustic model to discard audio data that the encoder determines is unlikely to be perceived. The amount discarded depends on the bitrate you choose. Lower bitrate means more data discarded, smaller file, lower quality. Higher bitrate means less data discarded, larger file, better quality.

MP3 bitrate tiers and what they are good for:

• 64 kbps: voice quality. Acceptable for speech in limited-bandwidth applications, like early podcast streaming or voice messages. Music sounds hollow, thin, and artifact-heavy. Not recommended for music under any circumstances.
• 96 kbps: better speech. Some music content may be passable if the music is not complex (solo voice, simple acoustic). Still clearly inferior to higher bitrates for music. You will notice the quality loss if you listen critically.
• 128 kbps: the old internet standard. Once considered acceptable for music. In 2026, most listeners can identify artifacts in 128 kbps MP3s on good headphones with complex music. Still fine for speech and casual listening on earbuds.
• 192 kbps: the sweet spot for most music. At 192 kbps, most listeners in most listening environments cannot distinguish an MP3 from a lossless file. This is what most music download services historically used and still use for MP3 distribution.
• 256 kbps: excellent quality. Perceptually transparent for virtually all music content in virtually all listening situations. iTunes Match and Apple Music use 256 kbps AAC (equivalent quality to about 320 kbps MP3).
• 320 kbps: the maximum standard MP3 bitrate. Mathematically, differences between 320 kbps MP3 and lossless exist, but in double-blind tests most listeners cannot identify them. This is the "audiophile" MP3 tier. Spotify Premium streams at this quality.

Variable bitrate (VBR) MP3 is an alternative to constant bitrate (CBR). VBR uses higher bitrate for complex sections and lower bitrate for simple sections, targeting a quality level rather than a fixed bitrate. VBR quality levels in the LAME encoder go from -V 0 (highest quality, roughly equivalent to 220 to 260 kbps CBR) to -V 9 (lowest). VBR -V 2 (roughly equivalent to 190 kbps) is the recommendation for general music use. VBR tends to produce smaller files than CBR at equivalent perceptual quality.

MP3's greatest strength is compatibility. Every device made in the last 30 years plays MP3. Car stereos, ancient MP3 players, Android, iOS, Windows, Mac, Linux, every game console, every smart speaker. If your priority is that someone can play the file on literally any device, MP3 is your format.

M4A/AAC Explained

M4A is an MPEG-4 audio file container. The audio codec inside it is AAC (Advanced Audio Coding). Sometimes people say M4A, sometimes they say AAC. These terms are used interchangeably in casual conversation, though technically AAC is the codec and M4A is the file format. Like saying MP3 when you mean an .mp3 file.

AAC was designed as the successor to MP3 and is objectively better at the same bitrate. Not slightly better. Noticeably better. At 128 kbps, AAC sounds closer to 192 kbps MP3. At 256 kbps, AAC is perceptually transparent for virtually all music. This is why Apple, YouTube, and most streaming services use AAC rather than MP3 as their standard codec.

The quality advantage is roughly 20 to 30% efficiency gain. That means a 128 kbps AAC file has perceptually similar quality to a 160 to 192 kbps MP3 file, at a smaller file size. Or put differently: at the same file size, AAC sounds better than MP3.

Compatibility in 2026 is excellent. iOS has always supported M4A natively. Android has supported it since Android 3.0. Windows Media Player, VLC, iTunes, Spotify, Apple Music, YouTube, and virtually every modern media player handles M4A/AAC without issues. The days of AAC being "an Apple-only thing" are well behind us.

AAC has two main encoding modes worth knowing:

• AAC-LC (Low Complexity): the standard mode used for music and general audio. This is what you get when someone says "AAC" without qualification. Good quality across all bitrates above 96 kbps.
• HE-AAC (High Efficiency AAC): designed for very low bitrates (below 64 kbps) where speech clarity matters more than music fidelity. Uses Spectral Band Replication to synthesize high-frequency content at lower bitrates. Used in internet radio, podcasting at low bandwidth, and voice applications. Sounds noticeably artificial at higher bitrates because the SBR synthesis is not as accurate as recording the real high-frequency content.

For music, use AAC-LC. For podcasts at normal bitrates, use AAC-LC. HE-AAC is only for situations where bandwidth is severely constrained.

The one real limitation of M4A: DRM-protected M4A files (.m4p) from early iTunes purchases cannot be played on non-Apple devices. DRM-free M4A files have no such restriction. Any M4A you create yourself is DRM-free. Any M4A you download from YTCut is DRM-free.

OGG/Vorbis Explained

OGG is a container format developed by the Xiph.Org Foundation. The most common audio codec used inside OGG containers is Vorbis, giving us the compound name OGG Vorbis. There is also OGG Opus, which uses the Opus codec (designed for speech and variable-latency applications), and OGG FLAC, which uses FLAC for lossless compression inside an OGG container.

OGG Vorbis was created specifically because MP3 was patent-encumbered at the time of its development. The goal was a free, open-source, patent-unencumbered audio codec that could match or beat MP3 quality. It succeeded. At similar bitrates, OGG Vorbis is comparable to or slightly better than MP3, though generally slightly behind AAC.

Quality ranking at equal bitrate: AAC is typically best, followed by OGG Vorbis, followed by MP3. The differences between OGG Vorbis and AAC are small and audio-content-dependent. At higher bitrates (192 kbps and above), all three are perceptually transparent and the ranking is academic.

The practical problem with OGG Vorbis is compatibility. It is not universally supported. iOS does not support OGG natively (though apps like VLC do). Many older car stereos and dedicated audio players do not support it. Web browsers generally support it for HTML5 audio, but even this varies.

Where OGG Vorbis shines is Linux environments, gaming (Steam supports it, many games use OGG for music and sound effects because of the patent freedom), and open-source software ecosystems. If you are distributing audio to a technically savvy audience that controls their own software stack, OGG Vorbis is a perfectly reasonable choice. For general consumer distribution, stick with MP3 or M4A.

OGG Opus deserves a separate mention because it is genuinely impressive technology. Opus was designed for voice communication and low-latency applications. At 64 kbps, Opus sounds dramatically better than any other codec for speech content. At 96 kbps, it is perceptually transparent for voice. Discord uses Opus. Zoom uses Opus. WhatsApp voice messages use Opus. If you are working with voice recordings specifically and want the best quality at the lowest file size, Opus is worth investigating. Compatibility is still limited compared to MP3 and AAC, but growing.

FLAC Explained

FLAC (Free Lossless Audio Codec) is what you get when you want the lossless quality of WAV but at a more manageable file size. FLAC compresses audio data without discarding any of it. Decompressing a FLAC file gives you bit-for-bit identical audio to the original uncompressed source.

Typical compression: FLAC reduces file size by 40 to 60% compared to WAV, depending on the audio content. A 31 MB WAV file of a three-minute song becomes roughly 15 to 20 MB as FLAC. Still much larger than MP3 or AAC, but much smaller than WAV while retaining every bit of the original data.

FLAC supports extensive metadata. Album art, track titles, artist names, composers, custom tags. This makes it suitable for music archiving in a way that WAV is not.

Compatibility has improved significantly. Android supports FLAC natively. iOS has supported FLAC since iOS 11. VLC, Foobar2000, Winamp, and most modern media players handle FLAC without plugins. Spotify and Apple Music now deliver some content in lossless quality, using FLAC and ALAC respectively. The ecosystem has caught up to where FLAC is no longer an enthusiast-only format.

What FLAC is not: a production format. Your DAW should still record to WAV or AIFF during sessions because FLAC encoding and decoding adds tiny amounts of latency and processing overhead that matter in a real-time recording context. FLAC is an archiving and distribution format for lossless content.

FLAC is also patent-free and open-source, which is why it has become the standard lossless format for the open-source ecosystem. Apple uses ALAC (Apple Lossless Audio Codec) instead, which is also open-source since 2011 but less universally supported outside Apple's ecosystem.

File Size Comparison Table

All sizes are approximate for a 3-minute stereo audio track. Actual sizes vary based on audio content complexity.

The takeaway from this table: FLAC is roughly half the size of WAV while being lossless. M4A/AAC gives you better quality than MP3 at the same file size. And the difference between a 128 kbps file and a 320 kbps file is about 4 MB for a three-minute song, which is nothing in 2026 terms but mattered a great deal when people had 20 GB iPods.

Quality at Equal Bitrate

This is the question everyone wants answered: at the same bitrate, which format sounds best?

The general ranking for music quality at equal bitrate: AAC is best, OGG Vorbis is second, MP3 is third. This ranking is well-established by listening tests including the hydrogenaudio.org blind tests that have been running for over two decades.

However, important caveats apply:

At 192 kbps and above, all three are effectively transparent for virtually all listeners in virtually all conditions. The ranking is meaningful at lower bitrates where codec efficiency determines whether you can hear artifacts. At 256 or 320 kbps, debating which codec is "better" is mostly a theoretical exercise. None of them are going to sound bad.

The quality of the encoder matters as much as the format. A poorly tuned AAC encoder can sound worse than a well-tuned MP3 encoder. The LAME MP3 encoder is exceptionally good and is what most software uses. Apple's AAC encoder (CoreAudio AAC) is excellent. Fraunhofer's FDK AAC is excellent. A random cheap AAC encoder from some obscure software may sound worse than LAME at the same bitrate. Format quality in practice depends on which specific encoder implementation you are using.

Content type affects the ranking. OGG Vorbis handles some types of acoustic music exceptionally well. AAC has an edge with electronic music and complex frequency content. These are generalizations, not rules. The differences at normal bitrates are small.

Your playback hardware matters. The quality difference between 128 kbps AAC and 320 kbps MP3 is inaudible on basic earbuds. On good headphones or studio monitors, it becomes apparent. The format debate is most relevant for quality-conscious listeners with quality playback equipment.

The Generation Loss Problem

This is the mistake that ruins audio quality and is entirely avoidable once you know about it. Generation loss, also called generational loss, is what happens when you encode a lossy file into another lossy format.

Here is the scenario. You download an MP3 at 192 kbps. You import it into your audio editor. You make some cuts, add some music, adjust the levels. You export as MP3 at 192 kbps. Then you realize you need to make one more change, so you import the exported MP3, edit it, and export again as MP3.

Every time you open a lossy file, decode it to edit it, and then re-encode it as lossy again, you introduce a new round of quality loss. The psychoacoustic model makes decisions about what to discard based on the current file. But the current file already has artifacts from the previous encode. Those artifacts interact with the new encode's decisions. Over multiple generations, the quality degrades noticeably. Voices can develop a watery, flanging quality. High frequencies can disappear. Transients (sharp attack sounds like snare hits or consonants) can develop pre-echo.

The fix is simple: always edit from the highest-quality source you have, and export lossless (WAV or FLAC) for intermediate versions until you are ready for your final export.

If you are editing audio that you downloaded as MP3, you are already starting with generation 1 of loss. You cannot recover the data that was discarded when the MP3 was created. What you can do is avoid adding more loss. Export intermediate edits as WAV. Only convert to MP3 or AAC as the final step for delivery.

If you are working from your own recorded audio, record to WAV. Edit in WAV. Make every version of the project in WAV. Only at the very end, when the file is done and you are ready to distribute it, export to MP3 or M4A.

One generation of lossy encoding at a good bitrate is fine. Many generations are not. Treat lossless formats as your working format and lossy formats as your delivery format. This single principle protects audio quality more than almost anything else.

Decision Framework by Use Case

Enough theory. Here is the answer to "which format should I use" for the situations you are actually likely to be in.

Recording and editing master files

Use WAV. Your DAW should record directly to WAV (or AIFF, which is essentially equivalent). This is your master. It is the source of all future versions. Never compromise here. Storage is cheap. Your original performance is not reproducible.

Archiving music you want to keep indefinitely

Use FLAC. It is lossless, compressed to a reasonable size, supports full metadata including album art, and has excellent long-term compatibility. This is the audio equivalent of keeping the original files instead of just the JPEGs from a photo shoot. You might never notice the difference, but you will be glad you have them if you ever need to.

Maximum compatibility with any device

Use MP3 at 192 kbps or higher. If someone says "send me the audio file" and you do not know what software or device they will use to play it, send MP3. It plays everywhere. At 192 kbps, quality is excellent for most uses.

Best compressed quality when compatibility allows

Use M4A/AAC at 256 kbps. Better sound quality than MP3 at the same or lower bitrate, excellent compatibility on modern devices and platforms, supported by every major streaming service. This is the format Apple uses for its entire music library for good reason.

Podcast distribution

Use MP3 at 128 to 192 kbps. Podcast apps are built around MP3. Some support M4A, but MP3 is the universal standard. For voice-only podcasts, 128 kbps MP3 is adequate. For music-heavy podcasts or audiophile listeners, 192 kbps. Do not use FLAC or WAV for podcast delivery. The file sizes are unreasonable for what is primarily a bandwidth-constrained medium.

Music production for streaming platforms

Deliver WAV to the distributor. Distrokid, TuneCore, CD Baby, and similar distributors require WAV or FLAC for submission. They handle the conversion to the formats each streaming platform uses. Never send MP3 to a music distributor. They will accept it but the re-encoding from MP3 to streaming formats introduces generation loss. Send lossless.

Podcast with video (YouTube, Vimeo)

Export audio as AAC 192 to 256 kbps embedded in MP4. YouTube and Vimeo both use AAC internally. Sending them a video with AAC audio is the most efficient option. Sending MP3 audio in the video container causes a re-encode to AAC on upload.

Apple-specific ecosystem (GarageBand, Logic, Apple podcasting)

Use M4A. Apple's tools export M4A by default. It is their native format. There is no reason to convert to MP3 if you are staying within the Apple ecosystem. If you are handing off to someone else, ask what they need before converting.

Gaming and open-source projects

Use OGG Vorbis. Games use OGG extensively because it is free of licensing concerns, well-supported by game audio engines (FMOD, Wwise, Unity, Unreal all support OGG natively), and performs well at moderate bitrates. The limited consumer device compatibility is irrelevant when the game engine is handling playback.

Voice-only content where bandwidth is severely constrained

Use OGG Opus at 64 to 96 kbps. At these bitrates, Opus sounds dramatically better than any other codec for voice. Discord, Zoom, and VoIP applications use Opus for exactly this reason. If you are streaming a talk radio show or voice podcast at low bandwidth, Opus at 64 kbps sounds better than MP3 at 128 kbps for voice.

How YTCut Handles Audio Formats

When you download audio from a YouTube video using YTCut, here is what you are actually getting.

MP3 download: YTCut extracts the audio from the YouTube video and encodes it to MP3 at 192 kbps. This is a deliberate choice. 192 kbps is the widely-accepted sweet spot for music quality. It sounds excellent, the file sizes are reasonable, and compatibility is universal. You are not getting a re-encode of a re-encode here. The process goes from YouTube's source audio (which is AAC at 128 kbps or higher depending on the video's quality level) to 192 kbps MP3. One encoding step.

M4A download: YTCut can also provide the audio in M4A/AAC format. YouTube videos are already encoded with AAC internally. In many cases, YTCut can extract the AAC audio stream directly without re-encoding it, which means zero quality loss from the extraction step. The file you get is exactly the audio as YouTube stored it. This is actually the highest-quality download option when your device supports M4A.

WAV download: For users who need uncompressed audio (editing, production work), YTCut can export as WAV PCM 16-bit. This decodes the YouTube audio stream to raw PCM. The file will be larger than the MP3 or M4A, but there is no lossy encoding step after the initial extraction. Note that the source is still YouTube's AAC stream, so the WAV will not have more information than the AAC had. You are getting the audio from a lossless container, but the content itself was encoded as AAC at some point in its history. Still useful for editing workflows where you want to avoid adding another generation of lossy encoding on top.

The format recommendation for YTCut downloads: if you want the audio for casual listening, MP3 at 192 kbps is great. If you want the highest-quality download and your device supports it, choose M4A. If you need the audio for editing in a DAW or audio production context, choose WAV and keep that as your source master for any further editing.

Bitrate Guide: When Is 128 kbps Enough?

The bitrate debate has been going on since MP3s existed. Here is a clear-headed take based on what actually matters in 2026.

128 kbps is enough for:

• Podcast audio listened to on earbuds during a commute
• Audiobooks
• Talk radio streams
• Voice messages
• Background music in a video where the audio is compressed further by the video platform anyway
• Any speech content where you are not expecting critical listening

128 kbps is not enough for:

• Music listened to on good headphones by anyone who cares about audio quality
• Music production source files
• Audio that will be re-encoded (generation loss problem)
• Classical music with wide dynamic range and complex instrumentation
• Electronic music with lots of high-frequency content and sharp transients
• Any situation where the listener might compare it to a lossless reference

192 kbps is enough for: virtually all music listening on virtually all consumer headphones and speakers. The vast majority of the listening public, including many self-identified audiophiles, cannot reliably distinguish 192 kbps MP3 from lossless in double-blind tests. This is the format recommendation for music distribution when lossless is not practical.

320 kbps is for: listeners who want the absolute maximum from a lossy format, archiving purposes where you might not have access to the lossless original, situations where you know the file will be re-encoded somewhere downstream and you want to start with the best possible quality, and peace of mind for people who are not confident in blind tests but still want to know they have the best quality available.

The honest truth: for most people in most situations, 192 kbps MP3 or 128 kbps AAC is all they need and they will never hear the difference from anything higher. The upgrade to 320 kbps is a marginal improvement that requires good equipment and critical listening to perceive. There is no shame in choosing the format that is practically sufficient rather than theoretically optimal. Your storage space and bandwidth can thank you.

FAQ

Can I convert MP3 to WAV to get better quality?

No. Converting MP3 to WAV makes the file much larger but does not add any quality. The WAV will contain exactly the same audio as the MP3, just in an uncompressed container. The data that was discarded when the MP3 was created is gone permanently. WAV cannot restore it. The only reason to convert MP3 to WAV is if you need to edit the file in a WAV-only workflow and want to avoid adding another round of lossy encoding. You are not gaining quality. You are preserving what you already have.

Which format does Spotify use?

Spotify uses OGG Vorbis internally for streaming, at 96 kbps (low), 160 kbps (normal), or 320 kbps (high quality/Premium). When you play Spotify on a web browser, it may use AAC instead due to browser codec support. Spotify Connect and the desktop app use OGG. Spotify's lossless tier (available in some markets) uses FLAC.

Does YouTube download quality depend on the format I choose?

The source material is the same regardless of output format. YouTube encodes audio at AAC 128 kbps for standard quality and up to AAC 256 kbps for higher-quality videos. When you use YTCut to download, the extraction quality is limited by what YouTube has stored, not by the output format you choose. Choosing MP3 320 kbps will not give you better audio than what YouTube's AAC 128 kbps source contained. It just means the encoding step targets a higher bitrate, which does not create quality that was not there.

Is AIFF the same as WAV?

Effectively yes, for most purposes. AIFF (Audio Interchange File Format) is Apple's version of uncompressed PCM audio. Both store the same raw audio data. AIFF is the default uncompressed format on macOS, WAV is the default on Windows. They are interchangeable in any modern audio software. Choose whichever one your workflow prefers. If you are entirely on macOS, AIFF. If you are cross-platform or on Windows, WAV.

What format should I use for music NFTs or digital art projects?

FLAC is the most appropriate for audio NFTs where the claim is delivering lossless audio. WAV is also common. M4A/AAC is used in some platforms' preview tracks. Whatever you choose, document the format and quality specifications clearly. "Lossless FLAC 24-bit 96 kHz" tells the buyer exactly what they are getting. "High quality audio file" tells them nothing useful.

My podcast host says to upload MP3. Can I use M4A instead?

It depends on the host. Most modern podcast hosts (Buzzsprout, Anchor/Spotify for Podcasters, Libsyn, Podbean) accept both MP3 and M4A. The RSS feeds they generate will usually point to the format you uploaded. Some very old podcast clients have issues with M4A. If universal compatibility is your priority, MP3 remains the safe choice for podcast distribution. If your audience is predominantly Apple users (iOS Podcasts app), M4A works perfectly.

What is the best format to send to a radio station?

Broadcast facilities almost universally want WAV at 16-bit 44.1 kHz (for standard broadcast) or 24-bit 48 kHz (for professional broadcast). Call ahead and confirm their technical requirements if you are not sure. Never send MP3 to a radio station without confirming they accept it. Many will not air MP3 because the quality after broadcast compression can be noticeably worse than broadcast-quality WAV.