SWF to AMR Converter

SWF to AMR — Is AMR the Right Target?

AMR (Adaptive Multi-Rate) is a 3GPP speech codec built for mobile voice calls and voice memos. It is extremely small and it is tuned for one job: intelligible human speech. This page pulls the embedded audio stream out of a legacy Adobe/Macromedia Flash (.swf) file and encodes it as AMR.

The short answer on suitability: if the Flash audio is narration, a tutorial voice-over, or an e-learning clip, AMR is a reasonable target and produces a tiny file. If it is music, a game soundtrack, or anything with stereo or high frequencies, AMR will hollow it out — it is mono-only and its sampling rates cap the reproducible bandwidth well below what music needs. In that case go to SWF to MP3 instead.

AMR Narrow Band vs AMR Wide Band

Both modes are offered on this page under Audio Codec, and the choice is not cosmetic — it changes the file's magic bytes, its sampling rate, and the entire bitrate ladder.

The bandwidth ceilings above are just Nyquist: an 8 kHz stream cannot carry content above 4 kHz, and a 16 kHz stream cannot carry content above 8 kHz. That is why a Flash soundtrack run through AMR-NB comes back sounding like a phone call — the top half of the music is not being thrown away by a quality setting, it is outside the format.

What the Codec Choice Locks Down

Picking a codec immediately constrains the other controls on the page, because AMR has no legal alternatives to constrain away:

This is deliberate. The underlying encoders reject anything else outright — feeding a raw 44.1 kHz stereo signal straight to an AMR-NB encoder fails rather than resampling itself — so the page removes the illegal choices before you can pick one, and the resample to mono at the required rate happens for you.

The Trim row is the exception you have to enforce yourself. Trimming is applied by seeking into the source before decoding, and a .swf cannot be seeked: the job logs "could not seek", encodes essentially nothing, and still hands you a file. We measured it on a working Flash soundtrack — a two-second trim produced a 38-byte .amr, against 9,574 bytes for the same clip converted whole. Leave Trim on "Unchanged", convert the whole clip, and cut the result afterwards.

How to Convert SWF to AMR

1. Upload Your SWF File: Drag and drop the .swf onto the page or click "Add Files" to browse. Several Flash files can be queued and encoded with the same settings.
2. Choose AMR Narrow Band or AMR Wide Band: Narrow Band writes an 8 kHz #!AMR file — the smallest option and the one old handsets expect. Wide Band writes a 16 kHz #!AMR-WB file that sounds noticeably clearer on speech.
3. Set File Compression: Leave it on Quality Preset, or switch to Constant Bitrate and pick a rung from the ladder for that codec — 4.75 to 12.2 kbps for narrowband, 6.60 to 23.85 kbps for wideband.
4. Convert and Download: Leave Trim on "Unchanged" — trimming a .swf during conversion produces an empty file, as explained above. Click Convert to download the .amr. Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically after a few hours — no sign-up, no watermark.

Does Your SWF Even Contain Audio?

This matters more for SWF than for any other source format on the site. An .swf is a compiled Flash program, not a plain media container. Our converter decodes the encoded media streams inside it — MP3, Flash ADPCM, Nellymoser and Speex on the audio side. It does not run the ActionScript and does not re-draw the vector timeline.

So a Flash file that wraps a real soundtrack extracts cleanly, while a Flash file that is only shapes, tweens and script has nothing to decode at all. Probing a vector-only SWF returns zero streams — the shape and timeline tags inside it are simply skipped — and the conversion fails rather than handing you a silent .amr. The same applies to a SWF that carries video but no audio track: with no audio stream present there is no input for an audio target, so the job stops.

Banner ads, logo stings, navigation widgets and most Flash games fall into that second group. Files built as Flash video players, animated shorts with a music bed, and narrated courseware fall into the first.

Frequently Asked Questions

Will a Flash music soundtrack survive as AMR?

Not in any recognisable form. AMR is mono and speech-optimised; narrowband tops out below 4 kHz of audio bandwidth and wideband below 8 kHz. Cymbals, string harmonics, sibilance and the whole stereo image are outside what the format can represent, so a song comes back thin and boxy. Convert music to SWF to MP3, which keeps stereo and full bandwidth at a still-modest size.

How much smaller is AMR than an uncompressed extraction?

Dramatically smaller. We measured the same six-second Flash soundtrack both ways: 1,064,502 bytes as uncompressed 44.1 kHz stereo PCM, and 9,702 bytes as AMR Narrow Band at the 12.2 kbps rung — a factor of about 110. That is the entire appeal of AMR, and also the reason it cannot be used for anything but voice.

Which bitrate rung should I pick for a voice clip?

For narrowband, 12.2 kbps is the top mode and the one to use unless you are squeezing a file for a very constrained device; 7.40 kbps is the classic "toll quality" compromise. For wideband, 12.65 kbps is the usual sweet spot, with 23.85 kbps available when clarity matters more than bytes. Below about 6 kbps on either codec, speech starts to sound synthetic.

Why can't I choose stereo or a different sample rate?

Because AMR does not have those modes. Both AMR-NB and AMR-WB are mono-only, and each writes exactly one sampling rate — 8000 Hz for narrowband, 16000 Hz for wideband. The page hides the illegal options rather than letting you select something the encoder would then reject, and it resamples your source down to the required rate automatically.

Will my .amr file play everywhere?

Not universally, and the codec you pick matters. AMR Narrow Band writes the #!AMR header that phones, voice-recorder apps and most players expect. AMR Wide Band writes a different header, #!AMR-WB, and older software that only knows narrowband will refuse it. If a player rejects your file, re-run the conversion on Narrow Band, or convert the result onward to a mainstream format.

Why did the conversion fail instead of giving me a silent file?

Because there was no audio stream inside the SWF to encode. That happens with pure vector animations, banner ads and most Flash games, and also with Flash movies that streamed their music from an external URL or a companion file at runtime. A silent output would require a real audio stream containing silence; when the container has no audio stream at all, the job stops. Check the folder the .swf came from — a loose .mp3 sitting next to it is usually the soundtrack.

Does this still work with Flash Player discontinued?

Yes, and no Flash runtime is involved. Adobe stopped supporting Flash Player on December 31, 2020 and blocked Flash content from running on January 12, 2021, but that ended playback, not the file. Any encoded audio stream inside the .swf is still readable and is decoded directly on our servers. If you want to see the animation rather than hear it, the open-source Ruffle emulator runs many legacy Flash files.

Can I extract just one line of narration?

Convert the whole clip first, then cut it. Do not use the Trim control on a .swf: trimming seeks into the source, the SWF reader cannot seek, and the result is a file that downloads but holds no usable audio — a two-second trim of a working soundtrack gave us 38 bytes, against 9,574 for the untrimmed conversion. Encode in full and then cut with the Audio Cutter, which accepts .amr directly.

The Flash file has lots of separate sound effects — do I get them all?

No. Flash stores each event sound as its own DefineSound object, and those appear as separate audio streams rather than one mixed track. Exactly one is selected for the output, so a game full of clicks and stings converts to a single short clip. Files built around one continuous streaming soundtrack — narration, a music bed — are the ones that behave as expected.

Should I use AMR at all, or a different format?

Use AMR when the audio is speech, the destination is a phone or a voice-memo workflow, and size is the constraint. Use SWF to MP3 for anything you want to actually listen to. If you need an uncompressed editing master instead, SWF to AIFF writes 16-bit PCM. The full list of targets is on the SWF converter hub.