Merge Audio

The most important part of an audio merge is usually not the final download button. It is deciding what remains in each clip and what the listener should hear when one source becomes the next. This page keeps those choices visible instead of joining every complete file with an unexplained default.

How to merge audio files

1. Choose two or more files. Add up to ten files with a combined source size no larger than 100 MB and a combined source duration no longer than 15 minutes. MP3 and WAV are reliable starting points; other formats work only when the current browser can decode their container and codec.
2. Put the clips in playback order. Drag the grip, or use the up and down buttons. The numbered list is the exact order used for the generated file.
3. Trim each source. Enter start and end timecodes to remove an introduction, silence, an unwanted ending or any other material outside the part you need.
4. Preview each selected range. The preview begins at that clip's start time and stops at its end time. Listen for clipped words, incomplete notes and unnecessary silence.
5. Set clip volume and boundaries. Adjust individual volume, optionally peak-match the clips, and choose a direct join, silent gap or crossfade.
6. Choose one output format. Select MP3 or WAV, 44.1 kHz or 48 kHz, and stereo or mono. MP3 also provides 128, 192, 256 and 320 kbps settings.
7. Merge and inspect. Listen across every boundary and read the result panel before downloading the new file.

Any edit to order, trim, volume, transition or output settings clears the previous result. This prevents an older download from being mistaken for the arrangement currently shown on the page.

Arrange and trim every clip

File selection order is not always playback order. A browser or operating system may return several selected files in an order that differs from the story, interview or track sequence you intended. Use the visible numbering as the source of truth.

Drag-and-drop is convenient on desktop, but the up and down buttons provide the same operation without precise pointer movement. On a touch device, the buttons are usually the more predictable choice.

Each clip has its own start, end and volume controls. Time fields accept seconds, M:SS.mmm or H:MM:SS.mmm. A minimum selected length of 0.05 seconds prevents an empty section from entering the timeline.

The small waveform is an overview rather than a full editor. Darker bars represent the selected portion and lighter bars show audio that will be excluded. Use it to notice obvious silence or activity, then rely on preview playback and timecodes for the final decision.

If one source needs detailed sample-level cutting, prepare it first with Audio Cutter. Merge Audio is designed for ordering and joining several ranges, not for editing many internal regions from the same file.

Choose the right audio transition

The same clips can feel very different depending on what occurs at their boundaries.

Direct join

A direct join places the next selected range immediately after the previous one. No extra silence is inserted and the clips do not overlap. This is useful for consecutive spoken sections, split recordings from the same event or clips that already contain the timing they need.

Direct joins can expose a sudden waveform or background-noise change. Preview the merged result across every boundary. If the transition clicks or feels abrupt, trim a little more carefully or use a short crossfade.

Silent gap

A silent gap inserts the selected duration between every pair of clips. The output becomes longer by that amount for each boundary. A gap can separate chapters, questions, announcements or practice prompts without requiring a separate silent source file.

Use a short gap when the listener only needs a breath. A long gap may feel like playback stopped, particularly when the sources already contain trailing silence.

Crossfade

A crossfade overlaps the end of one clip with the beginning of the next. This page applies complementary equal-power curves: the outgoing clip falls while the incoming clip rises across the same overlap.

The requested crossfade is capped at half the duration of either adjacent selected clip. That rule prevents a short source from being completely consumed by the transition. The result report shows the actual range of boundary durations when one or more short clips required a smaller overlap.

Crossfades work well when two music or ambience clips should hand off smoothly. They can be inappropriate for speech when overlapping words would reduce clarity. A 0.25–0.75 second crossfade is a practical starting point for many boundaries, but the correct value depends on the actual material.

This is a concrete advantage of the page's result report: the transition control states what you requested, while the generated result states what was actually possible after the clip lengths were known.

Manage volume without confusing peak and loudness

Every clip has a 0–150% volume slider. Use it after previewing adjacent clips, especially when recordings came from different microphones, rooms or applications.

The optional Peak-match clips control measures the highest absolute sample in each selected and resampled clip. It adjusts that peak toward -1 dBFS before applying the clip's volume slider. Amplification is capped at 4× so a nearly silent recording is not increased without limit.

Peak matching is not loudness normalization. Two clips can reach the same peak and still sound different because perceived loudness depends on the level and frequency content across time, not only the single highest sample. The tool states this limitation beside the control rather than presenting peak matching as a complete mastering process.

Use peak matching as a starting point when source levels differ substantially. Then listen to consecutive clips and correct the volume sliders by ear. For broadcast or platform loudness targets, use a dedicated loudness meter and editor that reports standards such as LUFS; this page does not calculate LUFS.

Crossfades and raised clip volumes can add overlapping samples above full scale. The mixer checks the combined peak before encoding. If it exceeds full scale, the complete output receives enough safety attenuation to bring that peak to approximately -0.09 dBFS. The result panel reports whether this safeguard was applied.

Choose MP3 or WAV output

All selected ranges must share one sample rate and channel layout before they can become one timeline. This tool resamples every clip to 44.1 kHz or 48 kHz and converts it to stereo or mono according to the output controls.

MP3

MP3 creates the smaller delivery file and is newly encoded at 128, 192, 256 or 320 kbps. A practical starting point is 192 kbps for mixed general-purpose audio. Use a higher setting when music detail matters and file size is less restrictive.

Merging existing MP3 files is not a lossless concatenation in this implementation. Each file is decoded, processed and encoded again so trimming, resampling, volume changes and crossfades can operate on audio samples. A high bitrate cannot restore information already missing from a source.

WAV

WAV output uses uncompressed 16-bit PCM at the selected sample rate. It is usually much larger than MP3 and is useful when another editing step follows or when you want to avoid MP3 compression in the merged result.

WAV does not improve a low-quality source. It also does not preserve each source's original bit depth; this browser encoder specifically creates 16-bit PCM.

Sample rate and channels

Choose 44.1 kHz for a common music-oriented delivery rate and 48 kHz when the result is heading into many video workflows. Choosing the larger number does not add missing detail to a lower-rate source.

Stereo keeps two output channels. A mono source is copied into both stereo channels, while a source with more than two decoded channels contributes its first two channels. Mono averages the decoded source channels. This page does not provide surround mapping or individual channel routing.

The browser processing model is built on decoded audio buffers and offline rendering provided by the Web Audio API. Codec support can differ by browser, which is why the input description uses “browser-supported audio” rather than promising every file with a familiar extension.

Read the merged audio result

Start by playing a few seconds before and after every boundary. Confirm that the order is correct, words or notes are complete, silence feels intentional and crossfades do not overlap content that should remain separate.

Then read the report:

• Merged order: lists the source filenames in the exact generated sequence.
• Duration: reports the actual encoded timeline and the total selected duration before gaps or overlaps.
• Transition: distinguishes direct joins, silent gaps and crossfades. Variable crossfade durations are reported as a range.
• Peak: reports the highest generated sample in dBFS. It is not a LUFS loudness value.
• Safety attenuation: appears when overlapping or amplified clips would otherwise exceed full scale.
• Sample rate and channels: confirm the common format used for every source in the timeline.
• File size: is the actual generated Blob size, not an estimate based on the sources.
• Metadata: “Not copied” means source tags, artwork, chapters and embedded lyrics are not transferred.

Do not expect source file sizes to add up to the output size. Trimming, gaps, crossfades, resampling, channel conversion, format and MP3 bitrate all change the result.

Illustrative audio merge examples

Join three interview answers

Assume three MP3 files contain answers of 42, 58 and 35 seconds after trimming. Arrange them in question order, use a 0.40-second silent gap and set the quietest clip to 115% volume. The estimated output is 135.8 seconds before encoding: 135 seconds of selected audio plus two gaps.

Blend two music clips

Assume two stereo WAV clips contain selected ranges of 45 and 38 seconds. Choose a 2-second crossfade and 48 kHz stereo WAV output. The expected timeline is about 81 seconds because two seconds overlap. Listen through that overlap to confirm the rhythm and musical key do not clash.

Combine a mono voice with stereo music

Assume a mono introduction is followed by a stereo music clip. Choose stereo output. The mono introduction is present in both output channels, while the music keeps its first two decoded channels. Lower the music volume if the transition feels disproportionately loud.

Illustrative examples. Exact file size, peak, actual crossfade and browser compatibility depend on the selected sources and settings. Use the generated result panel for the actual output.

Scope and alternatives

base97 focuses on a transparent, single-pass merge: arrange clips, trim each source, preview the selected ranges, choose one boundary behavior, set a common output format and inspect the result. The output is built locally in a Web Worker after browser decoding and resampling.

Other established tools address overlapping but broader needs:

• 123apps Audio Joiner provides online track joining and crossfade as part of a larger audio-tool suite.
• Clideo Audio Joiner provides multiple-file ordering and crossfade within a cloud media workflow.
• Kapwing places audio merging inside a broader timeline editor suited to projects that also need visual content, layers or continued editing.
• Adobe Express connects audio combining with a broader content-creation environment.

Those alternatives may be a better fit for cloud projects, account-based storage, visual timelines, collaboration, video composition or many layers. The base97 difference is narrower and verifiable: per-clip trim and volume controls, explicit transition math, peak matching labeled honestly, automatic clipping protection and a result report that exposes the actual timeline.

This version does not support URL imports, cloud drives, batch output, more than ten sources, multi-track layering, independent transition settings per boundary, waveform dragging for trim points, recording, AI silence removal, beat matching, tempo matching, key detection, speech transcription, LUFS normalization, surround routing, metadata copying or project saving.

Troubleshoot audio merging

One file will not open

A familiar extension does not guarantee that the current browser can decode the codec inside it. Try MP3 or WAV, or convert the source with a trusted desktop editor first. The complete selection must also stay within the 100 MB and 15-minute page limits.

The clips play in the wrong order

Read the numbered list from top to bottom. Move a clip with the drag grip or arrow buttons, then generate a new result. The previous result is cleared whenever the order changes.

A word or note is cut off

Extend that clip's start or end time and use Preview cut again. The small waveform helps locate activity but does not replace listening.

The crossfade overlaps speech

Reduce the duration or choose a direct join or silent gap. A crossfade is an overlap, so both clips are audible at the same time.

The volume still sounds uneven after peak matching

This is expected when clips have different perceived loudness despite similar peaks. Adjust the individual volume sliders by ear or use a dedicated LUFS-based workflow when a formal loudness target is required.

The result is quieter than expected

The result panel may report safety attenuation. This happens when clip gains or overlap would otherwise push the combined samples above full scale. Lower the loudest clip or reduce overlapping material if you want to avoid that output-wide adjustment.

Processing fails or uses too much memory

The browser holds decoded sources, resampled ranges and the merged timeline in memory. Remove unused clips, trim more aggressively, choose mono output or use a desktop editor for long production recordings.

Merge Audio FAQ