Initializing... drag & drop files here
Supports: FLV
This tool pulls the soundtrack out of an FLV (Flash Video) file, discards the picture, and re-encodes the audio as AMR — a tiny, mono, narrowband speech codec. That makes it the right choice only when the audio is talking — a recorded lecture, voice memo, interview, sermon, or any spoken-word Flash-era clip you want to shrink for an old phone or a voice archive. If your FLV holds music or a rich soundtrack, AMR is the wrong target and this page will tell you where to go instead.
AMR (Adaptive Multi-Rate) was set as the standard speech codec by 3GPP in 1999 for GSM and UMTS mobile voice calls — phone calls, .amr voice recordings, and MMS clips. AMR-NB (narrowband) samples at 8 kHz and runs at 4.75–12.2 kbps, mono only, capturing roughly the 300 Hz–3.4 kHz telephone band. It uses an ACELP algorithm that models the human vocal tract and throws away everything outside the speech band.
That design is why music comes out badly: most musical content lives above 3.4 kHz (cymbals, strings, harmonics), and AMR simply discards it — songs turn muffled, warbly, and telephone-quality, with percussion reduced to mush. There is also a double quality hit: FLV audio is already lossy (typically MP3 or AAC), so encoding it into a low-bitrate speech codec re-compresses an already-compressed stream. The result is a big quality drop in exchange for a very small file. Use AMR here only if the source is pure speech. For anything musical, convert to FLV to MP3 instead — full-spectrum stereo at a still-small size.
.flv files. Only the audio track is extracted — the video is discarded. Batch upload is supported.The Constant Bitrate dropdown is the one setting that matters for speech quality. AMR-NB defines eight fixed modes — 4.75, 5.15, 5.90, 6.70, 7.40, 7.95, 10.2, and 12.2 kbps — and every mode produces the same 8 kHz mono stream, just with more or fewer bits spent per second of audio. Higher modes sound clearer; lower modes make smaller files.
This conversion assumes a readable FLV with a speech soundtrack. If the FLV is corrupted or truncated (a common fate for files recovered from defunct Flash sites), the audio track may not decode — try repairing or re-downloading the source first. And if your goal is general-purpose audio rather than a telephony-grade voice file, AMR is simply the wrong tool: for music or mixed soundtracks convert to FLV to MP3, and to turn an existing AMR voice file back into something widely playable, see AMR to MP3.
No. AMR is a speech-only codec built around the ACELP algorithm — it models the human vocal tract and discards everything outside the ~300 Hz–3.4 kHz speech band. Music, sound effects, and ambient noise come out muffled and warbly, and percussion turns to mush. There is also a double quality hit: FLV audio is already lossy MP3 or AAC, and AMR re-encodes it again at a tiny speech bitrate. If your FLV has any musical content worth keeping, use FLV to MP3 instead. Pick AMR only when the source is pure speech.
FLV (Flash Video), introduced by Adobe in 2003, almost always carries MP3 or — in later files — AAC audio, both of which are lossy formats. Adobe ended Flash Player support on December 31, 2020, so nothing creates new FLV content today, but the container still decodes fine in VLC and ffmpeg, which is why extracting its audio still works. Because the source is already lossy, converting it to AMR stacks a second round of compression on top.
Two reductions stack. First, the video track is discarded entirely — typically the large majority of an FLV. Second, the remaining audio is re-encoded from a full-spectrum stereo MP3 or AAC stream down to a 4.75–12.2 kbps mono speech stream, another large drop. In our testing, a one-minute FLV clip of clear dialogue produced roughly a 90 KB AMR-NB file at 12.2 kbps — versus over 1 MB for the same minute as MP3.
Always mono, at 8 kHz — those are the only values AMR-NB defines. The encoder downmixes the FLV's stereo audio and resamples it to 8000 Hz automatically. This is fine for ringtones, MMS, IVR, and telephony, which are all mono pipelines. If you need stereo or a higher sample rate, AMR is the wrong format — use FLV to MP3 instead.
For a lecture or interview you actually want to hear clearly, pick 12.2 kbps — the top AMR-NB mode. For a voicemail-style archive where you only need the words to be understandable and the file as small as possible, drop to 5.90 or 4.75 kbps. Every mode produces the same 8 kHz mono stream; the bitrate just trades clarity against size.
Files are uploaded over an encrypted connection, processed on our servers, and deleted automatically a few hours after conversion — no sign-up, no watermark, never shared or made public. To go the other direction and turn an existing AMR voice file into a widely playable format, see AMR to MP3.