Converting video to audio sounds like a one-button task, but the right output depends on what you plan to do next. A podcast editor, a student saving a lecture, and someone sharing a short voice clip need different formats. This guide explains the complete workflow and the limits that responsible browser tools should disclose.
Quick workflow
- Choose the final video file from your device.
- Select MP3, WAV, M4A, or Original audio.
- Set start and end times if only one section is useful.
- Choose a bitrate for MP3 or M4A.
- Convert, download, and listen to the result before deleting the source.
With a browser-local converter, the video is processed by WebAssembly in the current tab rather than uploaded to a remote conversion queue. That distinction matters for interviews, meetings, client recordings, medical discussions, unpublished media, and personal clips.
Choose the output before you convert
MP3 is the compatibility choice. Nearly every player can open it, and 128 kbps is often adequate for speech. Use 192 kbps as a balanced default or 320 kbps when music is present and file size is less important. MP3 is still lossy at every bitrate.
WAV is the editing choice. The output is uncompressed PCM, which editors and transcription systems handle predictably. It avoids another lossy stage, but it cannot restore details already discarded by the AAC, MP3, or Opus audio inside the video. WAV files are large because size grows directly with duration, sample rate, bit depth, and channel count.
M4A usually contains AAC and offers strong quality for its size. It is a good choice for phones, podcasts, and efficient audio libraries. Like MP3, it re-encodes the audio and is lossy.
Original mode is different. It uses stream copy to preserve the encoded audio packets without re-encoding. Because videos can contain many codecs, the output uses an MKA container rather than pretending every source is MP3 or M4A. This is usually the fastest option and adds no generation loss, but MKA needs a capable player.
Trimming and browser performance
Trimming before conversion saves time and output space. Re-encoded formats can create samples around the requested range. Stream-copy trimming follows existing packet boundaries, so the first or last fraction of a second may not be exact.
Browser FFmpeg is slower than native desktop FFmpeg and can hold the engine, input, and output in memory together. A clear size limit protects phones and older computers from crashes. Close other heavy tabs, keep the page open, and use a desktop browser for long recordings. Multi-thread processing requires SharedArrayBuffer and cross-origin isolation; single-thread mode remains the compatibility baseline.
Verify the result
Play the beginning, middle, and end. Confirm that voices are present, the trim is correct, channels are intact, and the download opens in its intended app. Keep the original video until that check is complete. A successful progress bar proves that the command finished; listening proves that the output is useful.
How this guide maps to the converter
This is not a generic format checklist. The current converter accepts one local video up to 300 MB, maps its first audio stream, and runs ffmpeg.wasm in a worker inside the browser. Its live presets are MP3 at 128, 192, or 320 kbps; M4A at 128, 192, or 256 kbps; WAV as 48 kHz, 16-bit stereo PCM; and Original as stream copy into MKA. The worker is terminated after completion or cancellation so temporary processing files are released. Those implementation choices are why the guide recommends listening after conversion and retaining the source file.