Converting Video and Audio Formats Using FFmpeg

FFmpeg converts media by reading an input with -i, selecting or transforming its streams, and writing an output file. The correct command depends on whether you need a new container, new codecs, filters, smaller files, or compatibility with a particular player or editor. Start by identifying that requirement; the filename extension alone does not guarantee a usable result.

This guide covers stream copying (remuxing), full transcoding, video and audio examples, stream selection, quality and size controls, metadata, subtitles, automation, troubleshooting, performance, reliability, and cost. Commands assume a recent FFmpeg executable is installed and available as ffmpeg.

1. Understand containers, codecs, and streams

A container is the file format that holds media streams and related data. MP4, Matroska (MKV), WebM, MOV, AVI, and OGG are containers. A video stream may use H.264, H.265/HEVC, VP9, or AV1; an audio stream may use AAC, Opus, MP3, FLAC, or another codec. Subtitles, attachments, chapters, and data can also be present.

Changing a container is different from changing an encoded stream. A remux can put already encoded streams into another container without decoding them. Transcoding decodes and then encodes again. FFmpeg’s documentation defines transcoding as “the process of decoding a stream and then encoding it again.” Stream copying is faster and avoids another lossy generation, but only works when the destination container and target application accept the existing streams.

2. Inspect your FFmpeg build and the input

Support varies by version, compile-time options, and external libraries. The official FFmpeg format and codec reference describes general support, while your local build is authoritative for what you can run.

ffmpeg -version
ffmpeg -formats
ffmpeg -codecs
ffprobe -hide_banner input.mkv

ffprobe reports stream indexes, codecs, dimensions, frame rates, sample rates, channel layouts, durations, metadata, and disposition flags. Read its output before writing a conversion command when a file contains multiple tracks.

3. Choose streamcopy or transcoding

Approach	Command pattern	Use it when	Trade-offs
Streamcopy/remux	-c copy	Only the container must change and streams are compatible	Very little CPU; no quality loss; can fail for incompatible combinations
Transcode	-c:v ... -c:a ...	You need different codecs, filtering, resizing, deinterlacing, resampling, or mixing	More CPU and time; lossy encoders can reduce quality
Hybrid	Encode only selected streams, copy the rest	One stream needs conversion while another is already suitable	Requires deliberate stream mapping and compatibility checks

Remux without re-encoding

ffmpeg -i input.mkv -c copy output.mp4

This copies encoded streams into an MP4 container. It does not convert H.264 to another video codec or FLAC to AAC. If FFmpeg reports that a stream cannot be muxed into the destination, choose a compatible container or transcode that stream.

Transcode selected streams

ffmpeg -i input.mkv -c:v libx264 -c:a aac output.mp4

The video is decoded and encoded with x264, and the audio is encoded as AAC. A hybrid example keeps compatible audio untouched:

ffmpeg -i input.mkv -c:v libx264 -c:a copy output.mp4

Verify that the copied audio codec is accepted by MP4 and by the final player. FFmpeg does not validate every destination-specific compatibility rule for you.

4. Convert common video formats

AVI or MKV to broadly compatible MP4

ffmpeg -i input.avi -c:v libx264 -preset medium -crf .crf 23 -c:a aac -b:a 160k output.mp4

-crf controls constant-quality encoding for libx264; lower values generally mean higher quality and larger files. Treat 23 as a starting point, then inspect the output on the actual destination. The preset changes encoding speed versus compression efficiency; it does not define a universal quality level.

Convert to WebM

ffmpeg -i input.mp4 -c:v libvpx-vp9 -crf 32 -b:v 0 -c:a libopus -b:a 128k output.webm

Some builds may not include a requested encoder. Confirm with ffmpeg -encoders or ffmpeg -codecs.

Resize while converting

ffmpeg -i input.mp4 -vf "scale=1280:-2" -c:v libx264 -crf 23 -c:a aac output-720p.mp4

The -2 keeps the computed height even, which is accepted by many codecs. Filtering requires decoding and re-encoding; -c copy cannot be combined with a video filter.

Extract a short preview

ffmpeg -ss 00:01:00 -i input.mp4 -t 00:00:20 -c copy preview.mp4

Fast seeking before -i may begin at a nearby keyframe. For frame-accurate cuts, place -ss after -i and re-encode the affected streams.

5. Convert audio formats

Convert to MP3

ffmpeg -i input.wav -c:a libmp3lame -q:a 2 output.mp3

VBR quality values are encoder-specific. If a workflow requires a fixed bitrate, use -b:a 192k (or the bitrate required by that workflow) instead.

Convert to AAC in an M4A container

ffmpeg -i input.flac -c:a aac -b:a 192k output.m4a

Convert to lossless FLAC

ffmpeg -i input.wav -c:a flac output.flac

FLAC is lossless, but converting an already lossy MP3 back to FLAC cannot restore discarded information.

Resample or change channels

ffmpeg -i input.wav -ar 48000 -ac 2 -c:a pcm_s16le output.wav

-ar sets sample rate and -ac sets channel count. Downmixing changes the source’s channel arrangement, so listen to the result when phase or surround content matters.

6. Select streams deliberately

Inputs can contain multiple videos, languages, commentary tracks, subtitles, attachments, or data streams. Automatic selection may choose a different stream than you expect. Use -map with stream specifiers.

# First video and second audio stream
ffmpeg -i input.mkv -map 0:v:0 -map 0:a:1 -c:v libx264 -c:a aac output.mp4

# Copy all video and audio streams, omit data and attachments
ffmpeg -i input.mkv -map 0:v -map 0:a -c copy output.mkv

# Include the first subtitle stream
ffmpeg -i input.mkv -map 0:v:0 -map 0:a:0 -map 0:s:0 -c:v copy -c:a copy -c:s mov_text output.mp4

Indexes are zero-based. Add -map_metadata 0 to copy global metadata from the first input, or -map_chapters 0 to copy chapters. MP4 subtitle support depends on subtitle type; text subtitles often need conversion to a supported codec.

7. Control quality, size, and compatibility

• Quality: Use the chosen encoder’s quality controls such as CRF or a documented bitrate mode. Test against the destination rather than assuming one value fits every source.
• Frame rate: Preserve it unless the target requires a change. Use options such as -r carefully; forcing a rate can duplicate or drop frames.
• Pixel format: Consumer playback often expects a specific format. For H.264 compatibility, -pix_fmt yuv420p is commonly required by older devices, but verify the target.
• Fast start: For progressive MP4 playback, -movflags +faststart moves metadata toward the beginning of the file after encoding.
• Hardware acceleration: Builds may provide hardware encoders and decoders, but names and quality differ by operating system and GPU. Check ffmpeg -encoders, then validate output quality and driver behavior.

8. Batch conversion and safe automation

Use a separate output directory and never overwrite originals until a result has been checked. In a POSIX shell:

mkdir -p converted
for f in input/*; do
  base=$(basename "$f")
  stem=${base%.*}
  ffmpeg -nostdin -y -i "$f" -map 0:v:0 -map 0:a:0? \
    -c:v libx264 -crf 23 -c:a aac -b:a 160k "converted/${stem}.mp4"
done

The trailing ? makes the audio map optional, so silent videos do not fail solely because they lack an audio stream. For production jobs, capture exit codes, stderr, input duration, and output validation. Write to a temporary filename and rename only after FFmpeg exits successfully.

9. Troubleshooting common errors

Symptom	Likely cause	Fix
“Unknown encoder”	Your build lacks the requested encoder or external library	Check ffmpeg -encoders; install a build that includes it or choose an available encoder
“Could not find tag for codec”	The stream codec is not accepted by the output container	Choose another container, or transcode that stream with -c:v/-c:a
Output has the wrong language or commentary	Automatic stream selection chose another track	Inspect with ffprobe and add explicit -map
No audio in output	The input has no audio, mapping omitted it, or the codec is incompatible	Use -map 0:a?, inspect streams, and choose a supported audio codec
Large file or poor quality	Bitrate/quality settings do not match the source or target	Use the encoder’s quality mode, adjust CRF or bitrate, and compare on the intended device
Audio and video drift	Variable frame rate, timestamps, or a problematic source	Inspect timestamps; try a controlled transcode and avoid unnecessary streamcopy for damaged timing
“Invalid data found when processing input”	Wrong file, incomplete download, damaged media, or unsupported format	Run ffprobe, confirm the file is complete, and test another FFmpeg build
Command hangs in a script	FFmpeg is waiting for terminal input	Use -nostdin in noninteractive jobs and handle logs explicitly

10. Performance, reliability, and cost

Streamcopy avoids decode and encode work, so it is normally the least expensive operation in CPU time and storage churn. Transcoding cost depends on duration, resolution, filters, codec, preset, and hardware. Re-encoding also creates another lossy generation when the codec is lossy. Keep originals, measure representative files, and verify outputs with the actual player, editor, upload service, or device.

For reliable pipelines, pin the FFmpeg build, record ffmpeg -version, preserve stderr logs, set timeouts around the process, check its exit status, and validate that the output exists and has expected streams. Online FFmpeg documentation is regenerated nightly and follows the newest revision, so local help may differ; consult the documentation index and command reference alongside your installed build.

11. Or skip the browser setup

If your workflow also needs screenshots of web pages for documentation, visual regression, or media pipelines, ScreenshotNeo provides a single HTTP request instead of maintaining browser automation. Its API returns PNG, JPEG, WebP, or PDF. Cookie and consent banners, newsletter popups, and chat widgets are removed before capture; bot checks, blank pages, timeouts, failed loads, and cache hits are not billed. An MCP server provides take_screenshot, get_page_info, and capture_pdf tools for Claude, Cursor, and other MCP clients.

See the ScreenshotNeo API documentation for all options.

curl -G "https://api.screenshotneo.com/v1/shot" -d access_key=YOUR_API_KEY --data-urlencode url=https://stripe.com -o shot.webp

import requests
r = requests.get("https://api.screenshotneo.com/v1/shot", params={"access_key": "YOUR_API_KEY", "url": "https://stripe.com"}, timeout=90)
open("shot.webp", "wb").write(r.content)

const q = new URLSearchParams({ access_key: 'YOUR_API_KEY', url: 'https://stripe.com' });
const res = await fetch(`https://api.screenshotneo.com/v1/shot?${q}`);

Every response includes X-Page-Verdict and X-Billed headers so a caller can see what happened. The service supports full-page and element capture, device presets, custom viewports, retina scale, PDF controls, custom CSS and JavaScript, clicks, waits, request blocking, headers, cookies, user agents, timezone, geolocation, transparent backgrounds, resizing, caching TTLs, signed links, asynchronous jobs, webhooks, bulk capture, usage reporting, and an OpenAPI specification. The parameter names used by other screenshot APIs also work, which can simplify migration.

The Free plan includes 1,000 screenshots each month with no card. Paid plans start at $5 for 3,000 shots; yearly billing gives two months free. Create a free ScreenshotNeo account.

12. FAQ

Does changing an extension convert a file?

No. Use FFmpeg to remux or transcode, then verify the resulting streams and destination compatibility.

Can I use -c copy with filters?

No. Filters require decoded frames or samples, so the affected stream must be encoded again.

Why does the online command reference differ from my terminal?

The online documentation tracks a newer revision and is regenerated nightly. Check your local version and ffmpeg -h output.

Should I always convert to MP4?

Only when the target accepts the selected MP4 codecs and container features. WebM, MKV, MOV, or another format may be more appropriate for a specific player or editing workflow.

How do I preserve every stream?

Use explicit maps such as -map 0, then confirm that the destination container supports each video, audio, subtitle, attachment, and data stream.