How to Compress a Voice Recording or Lecture Audio to Send by Email
A phone records an hour-long lecture as a file in the hundreds of megabytes, and an inbox politely refuses anything a tenth of that size. Between those two numbers sits one of the most satisfying compression wins in all of media: spoken audio shrinks by 90–95% with no meaningful loss, because most of what a raw recording stores is detail no human voice ever produced.
The email attachment wall — and its hidden 33%
Most major mail providers cap attachments somewhere around 20–25 MB, and corporate mail servers are often stricter. The published number is also not the real number: email encodes attachments as base64 text, which inflates them by roughly a third in transit. A "25 MB" limit therefore accepts only about 18–19 MB of actual file, which is why attachments that look safely under the cap still bounce.
The practical target that survives every mainstream provider, with the overhead accounted for: keep the attachment under about 15 MB, and under 10 MB if you are sending to an unknown corporate server. Those numbers sound tight for an hour of audio — until you do the speech math.
Why speech is the easiest audio to compress
A human voice occupies a narrow slice of the audible spectrum — the information that makes speech intelligible lives almost entirely below 8 kHz, a fraction of what music uses. A raw recording at 44.1 kHz stereo is storing frequencies a voice never produced, in a second channel that duplicates the first. Compression for speech is mostly the act of not storing what was never there.
That is why 64 kbps, mono, at a 32 kHz sample rate — the Smallest mode in our audio compressor — sounds essentially transparent on a lecture while cutting the file by an order of magnitude. Podcasts have shipped at this kind of rate for two decades. The trade-offs that make 64 kbps a poor choice for music, explained in our MP3 bitrate guide, simply do not apply to one person talking.
The arithmetic for an hour of lecture
File size is bitrate × duration, so the numbers are predictable to the megabyte. One hour at 64 kbps mono is about 28 MB; at 48 kbps about 21 MB; a 30-minute recording at 64 kbps is about 14 MB — comfortably inside any inbox. A full 90-minute lecture won’t fit under 15 MB at good quality, and honesty matters here: the tool tells you when a target is physically impossible instead of delivering mush.
For a hard cap, skip the mental arithmetic: the Target size mode takes the megabytes — say 15 — and computes the exact bitrate your recording’s duration allows, warning you if the combination cannot work. For a recording that genuinely cannot fit, the options are splitting it into parts, or sharing a link instead of an attachment.
What not to bother with
Zipping an audio file is the classic dead end: MP3, M4A and OGG are already entropy-coded, so an archive shaves a percent or two at best — the same reason making a ZIP smaller is hard when its contents are already compressed. Recording at a lower volume, trimming silence at the ends, or converting between lossy formats at the same bitrate likewise change little or nothing. Bitrate and duration are the only levers that move megabytes.
And if the recording is destined for a chat app rather than an inbox, the constraints are friendlier — see our guide to compressing audio for WhatsApp.
Step by step: from phone recording to sent email
Open the audio compressor and drop in the recording — phone memo formats (M4A), WAV from dedicated recorders, and everything else up to 500 MB all work. Choose Smallest for pure speech, or Target size with your provider’s cap minus a small safety margin. Compress, then use the preview players to spot-check a minute from the middle of the recording.
Download and attach. The file keeps its title and metadata, sends without tripping size limits, and the upload is automatically deleted from our servers within 24 hours. Total time for an hour-long lecture: a couple of minutes, most of it upload.
Step by step
- 01
Upload the recording
Drop the lecture or voice memo into the audio compressor — M4A phone recordings, WAV and MP3 all work up to 500 MB.
- 02
Pick Smallest or a target size
Smallest (64 kbps mono) suits any speech; Target size lets you name your provider’s cap directly and computes the bitrate for you.
- 03
Preview, download, attach
Spot-check the compressed audio with the preview player, download it, and attach — staying under about 15 MB clears every major provider.
Frequently asked questions
How big is one hour of recorded audio?
It depends entirely on the format: roughly 600 MB as WAV, 30–60 MB as a typical phone M4A recording, and about 28 MB at 64 kbps mono MP3 — the speech setting. Same hour, same words, an order of magnitude apart.
Is 64 kbps really enough for a lecture recording?
For speech, yes — comfortably. Voices stay clear and natural at 64 kbps mono; podcasts have used rates like this for decades. The setting only sounds bad on music, which uses far more of the frequency spectrum.
Should I zip an audio file to email it?
No — compressed audio formats barely shrink inside a ZIP (a percent or two at best), because their data is already entropy-coded. Re-encoding at a lower bitrate is the lever that actually moves megabytes.
What if my recording is still too big after compressing?
Physics has a floor: at some duration, no acceptable bitrate fits your cap. Split the recording into parts and send several emails, or upload it and email a link instead. The target-size mode tells you up front when a cap is impossible for the duration.