A dashcam card usually holds one long file, and the part that matters is a few seconds. The chain is: make the file playable, cut the segment, pull stills for the claim, compress the clip, then put the damage photos into one PDF. Everything runs on your device, so footage of a street, a plate or a passenger never leaves it.
Two rules set the order. Convert first only when the browser cannot play the original, because trimming a file that will not open is impossible. And trim before compressing, so every later step processes seconds instead of an hour. The one honest gap is proof: these tools change files, and nothing here can show that a file was not changed, so keep the original card copy untouched. Name the exported clips by date and time so they can be matched to a report later.
1. Make the file playable
Dashcam cards write in whatever container the manufacturer chose. Video Converter turns a file the browser can decode into MP4 with H.264 video and AAC audio, at CRF 23, with the index moved to the front so a phone can start playing before the download finishes. MP4 is the safest target for a claim portal, a chat app or an email attachment, because it opens on almost anything without extra software.
The hard limit is decoding, not conversion. If the browser cannot read the file, this page cannot repair it: some AVI variants, certain HEVC recordings and proprietary formats from cheap dashcams will fail to load. No tool here recovers a corrupted card either. In that case, export through the camera maker's app first, then convert the exported file here. The conversion runs locally, so the footage is not uploaded to a server.
Video Converter
Turn a dashcam file into MP4 with H.264 and AAC, ready for a phone or a portal.
2. Cut the thirty seconds that matter
Video Trimmer marks a start and an end on the timeline, or lets you play the clip and press Set start here and Set end here at the right moments. Fast mode copies the streams without re-encoding, so a long card file becomes a small clip in seconds and the pixels are identical to the original. The catch is that stream copy can only begin on a keyframe, so the clip may start a second or two before the point you picked.
Precise mode re-encodes and lands exactly on the chosen frames, which is worth the wait when the moment is the whole point, such as the second a light changes. One range per export, so several separate moments mean several passes. Trim before compressing, converting or extracting frames, because every later step then works on a few seconds instead of the whole file.
Video Trimmer
Keep the seconds that matter, in lossless copy mode or frame-accurate mode.
3. Pull stills for the claim
Extract Frames saves a still at a time you type or scrub to, accurate to a hundredth of a second, which is how you land on the frame where a plate, a signal or a lane marking is clearest. Interval mode takes one frame every N seconds, and count mode spreads a fixed number across the clip; both batch modes stop at 30 frames per run. PNG is lossless, while JPG and WebP are smaller with a quality slider.
What this cannot do is add detail that the sensor never captured. If a plate is smeared across a handful of pixels, pulling the frame at the perfect moment gives you the sharpest available version, not a readable one. The AI enhancer and upscaler on this site cannot invent characters either, and both upload the file. A practical habit is to save three or four frames around the moment, because motion blur and compression differ frame to frame and one may be readable where another is not.
Extract Frames
Save a still from an exact moment, or up to 30 across the clip, in PNG, JPG or WebP.
4. Compress to the claim form's limit
Claim portals and messaging apps often cap an attachment at a size a raw dashcam clip will never meet. Video Compressor exposes quality, resolution, frame rate and output format. For a claim, 720p at 30 fps is usually enough: a plate that is readable at all stays readable, and the file drops by 50 to 80 percent compared with the original. The progress bar and the log show exactly what the encoder is doing.
Two limits to plan around. There is no target size field, so you lower the quality slider and watch the output number rather than typing a cap. And compression is one-way, so keep the trimmed MP4 as the evidence copy and send the compressed one. Clips under 200 MB and a few minutes long work best, because the encode happens in browser memory.
Video Compressor
Set quality, resolution and frame rate, and watch the output size before you send.
5. Put the damage photos into one PDF
Phone photos of a bumper, a wing mirror and the other vehicle are usually three files of several megabytes each. Image Compressor takes the whole set, shrinks the files and can resize by percentage, with a before and after preview. Then Image to PDF turns them into pages, and the cards can be dragged into the order an adjuster will read: wide shot, close-up, plate, document. A4 or Letter, portrait or landscape, margin and fit are all selectable.
Finish with Compress PDF if the attachment is still too large. It rewrites the internal structure with qpdf instead of re-encoding the photos, so the images keep their quality; the tool is designed for files under 50 MB. The limit worth knowing is that compressing a PDF does not shrink a photo that is already poorly compressed, and it is not a scan optimiser. For photographs, the real saving happens in the previous step.
Image to PDF
Build one ordered PDF from the damage photos, with A4 or Letter pages.
6. Keep an audio copy of the exchange
Dashcam audio is often the clearest record of what was said at the roadside. Video to MP3 extracts the soundtrack from the trimmed clip, and Audio Trimmer cuts the exchange down to the part that matters, with a fade of up to 5 seconds at each end. The trimmer opens MP3, WAV, M4A, AAC, OGG, FLAC and Opus and exports MP3 or WAV, and it handles one region per file, so several exchanges mean several passes.
If you need to add your own account of the incident, Voice Recorder captures it in the browser, and the recording can be converted to MP3 before download. Nothing is uploaded at any point. What these tools cannot do is transcribe the audio locally: the transcript tool on this site runs in the cloud and uploads the file, which is a real consideration when a passenger's voice is on the recording.
Audio Trimmer
Cut the roadside exchange out of the dashcam audio, with a fade at both ends.
What this chain cannot do
- No repair and no proprietary decoding. Some AVI variants, certain HEVC files and manufacturer-specific formats will not open in the browser.
- No detail recovery. A plate that is not legible in the source will not become legible after extraction, enhancement or upscaling.
- No timestamp or speed overlay. Nothing burns a clock, a date or a speed reading onto the frame.
- No proof of integrity. There is no hash, signature or timestamp, so the tools cannot demonstrate that a clip was not edited.
- No incident report. The output is files, not a form, a statement or a submission to an insurer.
- No target video size. Compression is tuned by quality and resolution, then checked against the output number.
- Compression is one-way. Always keep the original card file and the trimmed MP4 as the clean copies.
- No local transcription. The transcript tool uploads the recording to a server.