Troubleshooting · How we test and maintain this guide
Check the title, archive and extracted folder separately
A song title is a display label. A filename must also work as a file entry, avoid collisions and fit the chosen naming style. Some changes are deliberate: punctuation may be replaced, an empty title receives a fallback, and a repeated name needs a suffix. These are different from characters turning into unreadable text during extraction or spreadsheet import.
Keep the original ZIP while you investigate. Compare the name shown in TuneBundle, the filename field in tracks.csv, the entry name displayed by your archive program, and the actual extracted name. Record the first stage where they differ. If the workspace already shows the unexpected title, changing the ZIP extractor cannot restore a different source title.
Expected results from six controlled titles
| Input title | Title-only output |
|---|---|
| Same title | Same title.mp3 |
| same title | same title (2).mp3 |
| A/B: take? | A B take.mp3 |
| CON | _CON.mp3 |
| (blank) | Untitled.mp3 |
| 夜の音楽 | 夜の音楽.mp3 |
The title-only naming function normalises text, removes problematic path characters, handles reserved names and keeps Unicode text such as the Japanese example. Its collision comparison ignores letter case, so Same title and same title remain distinct files with a suffix. The suffix does not mean the second song is a duplicate recording. It only makes the output name unique in that prepared collection.
Title + ID gives you a stronger connection to the source identity. Number + title makes sequence visible in ordinary filename sorting. Neither choice rewrites embedded music tags, and neither establishes rights. The current naming rules operate on the prepared collection; an existing file in your local folder can cause the browser to assign another suffix during saving.
Recover readable names without altering the recording
- Extract a copy into a new empty directory. Do not overwrite the only copy while experimenting.
- Open the archive with an application that reads its Unicode filenames correctly. Compare the same entry in another extractor if the text is unreadable in only one program.
- If the CSV alone looks wrong, import it explicitly as UTF-8 with comma-separated columns. The exported CSV includes a UTF-8 byte-order mark.
- Match files using source ID, original filename, size and hash where available. Avoid matching solely on a damaged title.
- Rename copies only after establishing the mapping, then save a small old-name to new-name record beside them.
Do not change the extension to fix a title. Renaming a file from .m4a to .mp3 does not convert its audio, and it can make a player choose the wrong reader. Keep extension repair separate from human-readable naming. If you need a different audio format, prepare or convert a verified source and keep the relationship between the two versions.
Long folder paths and extra prefixes can also complicate a transfer. Test a short destination path with one affected file before renaming an entire library. This guide does not claim that every operating system, filesystem, cloud client and archive program accepts every Unicode name identically. The measured example establishes the tool's output names, not all downstream applications.
Retain identity when cleaning a collection
Preserve the untouched ZIP and its CSV until the renamed folder has been checked. If two songs share a title, include a take label or source identifier in your own naming plan. Do not remove a collision suffix from two different files and then merge them into the same directory without checking what would be overwritten.
A pure filename change does not change the audio bytes. In our separate local filing experiment, a renamed copy had the same file hash as its source. Writing metadata tags changed that hash even though decoded audio stayed the same. This distinction helps you diagnose whether only the label changed or whether a later editor also modified the file.
When requesting help, provide the original and displayed character strings as text, the naming style, extractor name and the point where corruption first appears. A screenshot can show the symptom, but the exact text makes a Unicode problem easier to reproduce. Remove private source links from any sample you share.