Initializing... drag & drop files here
Supports: FLV
Flash Video is the container that carried web audio and video through the 2000s, and Adobe ended Flash Player support on 31 December 2020, so .flv files today are archive material: lectures, screencasts, webcam captures, podcast rips and stream recordings. This page lifts the soundtrack out of one and re-encodes it to Opus, the IETF codec defined in RFC 6716. The video track is discarded — the output is an audio-only .opus file. Opus stays intelligible at bitrates where older codecs fall apart, which is exactly the right property for salvaged, already-compressed Flash audio.
| Property | Value |
|---|---|
| Standard | IETF RFC 6716 |
| Published | September 2012 |
| Designed by | Xiph.Org and Skype engineers, standardised through the IETF |
| Type | Lossy, royalty-free |
| Bitrate range | 6 kbps to 510 kbps |
| Sample rates | 8, 12, 16, 24 and 48 kHz — 44.1 kHz is not one of them |
| Internal design | A hybrid of SILK (speech) and CELT (music), switching or blending by bitrate and content |
| Where it is mandatory | WebRTC real-time audio; also the audio codec used by Discord voice |
| Native playback | Chrome, Firefox, Edge and Opera; Android; VLC and mpv; recent Apple systems |
| Weak spots | Older car head units, legacy portable players, and pre-Windows-10 systems without a decoder |
| Audio codec found in FLV files | Typically used for | Lossy? |
|---|---|---|
| MP3 | Most music and general web video | Yes |
| AAC | Later FLV files, especially H.264-era streams | Yes |
| Nellymoser Asao | Microphone and webcam captures recorded through Flash | Yes |
| Speex | Voice chat and low-bandwidth speech recordings | Yes |
| ADPCM | Older, simpler encodes | Yes |
| Linear PCM | Rare; uncompressed audio in an FLV wrapper | No |
.flv onto the page or click "+ Add Files" to browse. Several files can be queued and converted with the same settings in one pass.HH:MM:SS.sss..opus file. No sign-up, no watermark.| Content | Suggested bitrate | Channels | Notes |
|---|---|---|---|
| Voice, lecture, screencast | 24–48 kbps | Mono | Opus was designed for speech at these rates; files stay very small |
| Spoken word with background sound | 48–64 kbps | Mono or Stereo | Enough headroom for room tone and music beds |
| Mixed speech and music | 64–96 kbps | Stereo | A sensible default when you do not know what is in the file |
| Music from an old rip | 96–128 kbps | Stereo | Opus reaches transparency far lower than MP3 needs to |
| Extra margin before further editing | 160–192 kbps | Stereo | Only worth it if the FLV's own audio bitrate was high |
The one rule that beats all of these: do not set a bitrate far above what the FLV already contained. If the source was a 96 kbps MP3, encoding at 256 kbps produces a bigger file carrying exactly the same information.
Only the sound. Opus is an audio-only codec, so the Sorenson Spark, VP6 or H.264 video inside the FLV is decoded, discarded, and never written to the output — you get a .opus file containing the soundtrack alone. That is exactly what you want for rescuing a lecture, an interview or a music rip out of a Flash archive. If the picture matters too, convert FLV to MP4 instead.
Because Opus does not define a 44.1 kHz mode. RFC 6716 specifies the codec at 8, 12, 16, 24 and 48 kHz, so a 44.1 kHz source cannot be stored at its original rate. We snap the requested rate to the nearest one Opus actually supports, which for 44.1 kHz means 48 kHz — including when you explicitly pick 44100 Hz from the Audio Sample Rate list. Resampling is not audibly damaging at these rates, but it is worth knowing so the number in your file inspector does not come as a surprise.
A little, because this is a re-encode rather than a copy. FLV audio is almost always already lossy — MP3, AAC, Nellymoser Asao, Speex or ADPCM — so this is a lossy-to-lossy generation. Opus is far more efficient than the codecs Flash shipped with, particularly Nellymoser and Speex, so it can carry what remains at a smaller size, but it cannot undo loss that was baked in at record time. Match or slightly exceed the source bitrate and the added loss is inaudible for most listeners.
24–48 kbps in Mono is plenty. Speech was Opus's original design target — the SILK half of the codec came from Skype's speech work — and voice recordings from Flash-era screen captures were low-bandwidth to begin with. Downmixing to Mono is the single biggest saving on a talking-head recording, because a webcam or headset capture rarely holds real stereo information anyway.
Chrome, Firefox, Edge and Opera play Opus natively, as do Android, VLC, mpv and Discord, and recent Apple systems have added support. The gaps are the ones you would expect from a 2012 codec: older car head units, legacy portable players, and Windows installations older than 10 that need a decoder pack. If universal playback matters more than efficiency, convert FLV to MP3 instead — MP3 is worse per byte but plays on essentially anything with a speaker.
Yes. Set Trim to a Start time and a Duration, in seconds or HH:MM:SS.sss, and only that span is encoded. This is worth doing up front for long stream rips or multi-hour lecture captures, since it cuts both the processing and the size of the download. If you would rather cut the file after the fact, the Audio Cutter does the same job on the finished audio.
Because you are uploading the whole FLV, video track included, and the video is by far the larger part of it. A 200 MB Flash capture whose soundtrack is a 96 kbps MP3 still has to travel in full before the audio can be separated out — the resulting .opus may be a couple of megabytes. Nothing is wrong; the video simply has to arrive before it can be thrown away.
Not reliably. FLV stores loose metadata in an onMetaData script tag rather than in a standard tag format, and there is no direct mapping from that into the Ogg comment header an Opus file uses. Treat the output as untagged audio and add tags afterwards in a tag editor or your music library software if you need them.
Your FLV is uploaded over an encrypted connection, processed on our servers, and the .opus file is returned to you. Uploads and outputs are deleted automatically a few hours after conversion — no sign-up, no watermark, never shared or made public. The practical limit for a long Flash capture is upload size and time, since the file still carries its full video track on the way in.