OpenSpeaks Bento helps you organise audio and video files for language documentation projects. Load your files, choose a naming pattern, and download a ZIP with all files renamed. You can also check the total duration of your recordings and compress files to share more easily. Everything runs in your browser — no account or installation needed. How to use this tool →

📂

Drop files or a folder here

Or use the buttons below

Select Files Select Folder
💡 What to include for each recording
🎬One video file — the main recording (e.g. .mp4, .mov)
🎙️Separate audio, if recorded on a different device (e.g. .wav, .mp3)
📄Notes or annotation file (e.g. .eaf, .xml, .json)
💬Subtitles or transcript, if available (.srt, .vtt, .txt)
Naming Convention
For raw field recordings. Keeps names short and machine-readable.
Add sequence number (001, 002…)
Use lowercase file extension (.mp3 not .MP3)
40
If the full name is too long, a shorter version is used automatically.
Example name: kan_DG_20240601_elicitation_001.mp3
Rename Plan Preview
📋
Add files and click "Generate Rename Plan"
Files 0
📁
No files loaded
🎧

Drop audio or video files

To select multiple files: hold Ctrl (Windows) or Cmd (Mac) while clicking

Browse Files
ℹ️ MP3, WAV, MP4, WebM, and OGG files work in all browsers. MKV and FLAC files may not work in all browsers.
Progress
Waiting…0%
—
Audio total
—
Video total
—
Combined
0
Files
Duration of Each File
⏱
Load files to see durations
⚠️ This tool works in your web browser. For video files, a helper program (~30 MB) will be downloaded on first use — this needs an internet connection. Audio and image files work fully offline, with no download needed.
🗜️

Drop one media file

Audio, video, or image

Browse
Compression Settings
64 MB
Progress
Idle—
Result
💾
Compressed file will appear here
Things to know
  • Video: first use requires internet (~30 MB download, one time only)
  • Audio: works fully offline, no download needed
  • Images: works fully offline, very fast
  • Files larger than 1 GB may not work in the browser. Use a desktop app for very large files.
  • Video quality is reduced to reach your target size. Frame rate stays the same.
  • The compressed file is saved to your device.
🔍

Drop a media file to analyse

Audio or video — any format your browser can read

Browse
Options
Compute SHA-256 checksum (digital fingerprint for verifying the file has not been corrupted)
Measure audio loudness (may not work for all formats)
Progress
Idle—
Analysis Report
📊
Drop a file and click Analyse to see the report

What is OpenSpeaks Bento?

OpenSpeaks Bento is a free tool for people who work on language documentation projects. It helps you rename your recording files using a standard naming system, check the total length of your recordings, and make files smaller so they are easier to share.

Everything runs inside your web browser. You do not need to install anything, create an account, or have a fast internet connection (except for compressing video files for the first time).

📂 Organise

Rename your files using a standard pattern based on language, speaker, date, and category. Download a ZIP with all files renamed.

⏱ Duration

See the total length of your audio and video recordings. Useful for reporting and planning.

🗜️ Compress

Make your files smaller to share by email, WhatsApp, or Telegram without losing too much quality.

🔍 Analyse Media

Check your recordings for archival quality. See codec, resolution, loudness, and get recommendations for long-term preservation.

🔒 Private

Your files never leave your computer. All processing happens in your browser — nothing is uploaded to any server.

How to use the Organise tab

The Organise tab renames your files using a standard naming convention. It supports two modes: one for organising raw field recordings, and one for preparing files for upload to Wikimedia Commons or a community archive.

Step-by-step

1
Load your files. Drag and drop files or a folder into the drop area, or click Select Files to pick files one by one. Click Select Folder to load all files from a folder at once.
2
Choose a naming convention. Select Field recordings for raw working files, or Upload-ready for final files going to an archive or Wikimedia Commons. Then fill in the fields shown.
3
Click "Generate Rename Plan". You will see a preview of the new names. Old names are shown crossed out; new names are shown in green. Check that everything looks right before downloading.
4
Download the ZIP. Click Rename & Download ZIP. A dialog will ask if you want to compress the files first or download them as-is. Choose Download lossless ZIP to get all files renamed with no quality loss.
💡You can also export the rename plan as a CSV or TXT file — useful if you want to keep a record of what was renamed.

Field recordings convention

Use this for raw audio and video files from a recording session. Names are short and machine-readable.

  • Language code — A short code for the language being recorded. Use the ISO 639-3 standard (e.g. kan for Kannada, eng for English). Use the link next to the field to look up any language code.
  • Speaker name or initials — The name or initials of the person speaking (e.g. DG for Dinabandhu Gomango).
  • Date — The date the recording was made. Stored in the file name as YYYYMMDD (e.g. 20240601 for 1 June 2024).
  • Category — The type of recording: Elicitation, Narrative, Song, Interview, Wordlist, Transcription, or Other.
LANG _ SPEAKER _ DATE _ CATEGORY _ 001 . ext
kan_DG_20240601_narrative_001.mp4

If the full name is too long, a shorter version is used automatically:

kan-DG-20240601-001.mp4

Upload-ready convention

Use this for finished files being uploaded to Wikimedia Commons, an institutional archive, or a community repository. Names are longer and designed to be self-describing — anyone who sees the file name should be able to understand who recorded what, in which language, and about what topic.

  • Language code — ISO 639-3 code or the community's own standard code for the language (e.g. Thq for Eastern Tharu).
  • Language or community name — The full name of the language or project (e.g. Eastern Tharu, Kandha). This makes the file discoverable by language on Wikimedia Commons.
  • Knowledge holder / speaker — The full name of the community member whose knowledge or voice is being recorded. This is required for attribution and respects the speaker's right to be credited for their contribution.
  • Documenter / researcher — The full name of the person who conducted and recorded the session. Optional, but strongly recommended for projects where the researcher's identity is part of the record.
  • Topic — What the recording is about (e.g. Medicinal Plants, Harvest Songs). Use plain language rather than a category code.

Fields are separated by hyphens. When a documenter is provided:

CODE - Language name - Documenter - Speaker - Topic 01 . ext
Thq-Eastern Tharu-Sanjib Chaudhary-Achhai Chaudhary-Medicinal Plants 01.webm

Without a documenter:

Thq-Eastern Tharu-Achhai Chaudhary-Medicinal Plants 01.webm
ℹ️Hyphens are the field separator, so any hyphen within a field value (e.g. a hyphenated name) is automatically replaced with a space to avoid ambiguity.
⚠️The tool does not rename files on your computer. It creates a ZIP file containing copies of your files with the new names. Your original files are not changed.

How to use the Duration tab

The Duration tab reads the length of your audio and video files and shows you the total recording time. This is useful for reporting, grant applications, and planning transcription work.

Step-by-step

1
Load your files. If you already loaded files in the Organise tab, they will appear here automatically in the "Files from Organise" section. Otherwise, drag and drop files or click Browse Files.
2
Remove any files you don't need. Click the × button next to any file you want to leave out before measuring.
3
Click "Measure Duration" (for files from Organise) or wait for the tool to process files dropped directly into the drop area.
4
Read the results. The tab shows the duration of each file and totals for audio, video, and combined. Click Export CSV or Export TXT to save a report.
ℹ️MP3, WAV, MP4, WebM, and OGG files work in all browsers. MKV and FLAC files may not work in all browsers — if a file shows an error, try converting it to MP4 or MP3 first.

How to use the Compress tab

The Compress tab makes your files smaller so they are easier to send by email or messaging apps. It works for audio, video, and image files.

Step-by-step

1
Load a file. Drag and drop a file or click Browse. If you loaded files in the Organise tab, they appear in the "Files from Organise" section — click Load next to any file to select it.
2
Choose a size. Use the Quick settings dropdown to pick a common option (WhatsApp, Email, Telegram, Web review), or use the slider to set a custom target file size.
3
Click "Compress & Download". The compressed file is saved directly to your device.

What works offline and what needs the internet

🎬 Video files

The first time you compress a video, a helper program (~30 MB) is downloaded automatically. After that, it works without internet. You must open this tool via a web address (http:// or https://) for video compression to work.

🎙️ Audio files

Works fully offline. No download needed. Compressed audio is saved as WebM or OGG format.

🖼️ Image files

Works fully offline. Very fast. Compressed images are saved as JPEG or WebP.

📁 File size limit

Files larger than 1 GB may not work in the browser. For very large files, use a desktop application such as HandBrake (video) or Audacity (audio).

⚠️Compression reduces quality to make files smaller. For archiving, always keep the original uncompressed files and only compress copies for sharing.

How to use the Analyse Media tab

The Analyse Media tab examines your audio and video files and creates a detailed quality report for archival purposes. It helps you understand whether your recordings are suitable for long-term preservation and what steps you might take to protect them.

Step-by-step

1
Load a file. Drag and drop an audio or video file into the drop area, or click Browse to select one from your computer.
2
Choose options. You can enable or disable the SHA-256 checksum (a digital fingerprint used to verify your file has not been corrupted over time) and loudness measurement (checks whether audio volume meets broadcast standards).
3
Click "Analyse". The tool reads the file's metadata, computes the checksum, and measures loudness — all inside your browser. Nothing is uploaded to any server.
4
Read the report. Each parameter has a coloured dot: green means good for archival, orange means acceptable with caveats, and red means there may be a concern. Scroll down for specific recommendations.
5
Export. Save the report as Plain Text (for records), Wikicode (for Wikipedia and Wikimedia Commons pages), or JSON (for databases and tools).

What the report covers

File details

Format, size, duration, and overall bitrate. These tell you the basic properties of your recording file.

Source device

The tool tries to identify what camera or phone made the recording. Different devices have different archival characteristics.

Video quality

Resolution, codec, colour depth, chroma subsampling, frame rate, and whether the video uses lossless or lossy compression.

Audio quality & loudness

Codec, sample rate, channels, loudness measurement (EBU R128), and compliance with broadcast standards (Netflix, EBU, ATSC).

🔒Your files never leave your device. The analysis runs entirely in your browser using a tool called MediaInfo.js. No data is uploaded to any server.
💡The SHA-256 checksum is a unique digital fingerprint of your file. Store it alongside your archival files (for example, in a small text file called recording.sha256). If you re-check the file months or years later and the fingerprint has changed, it means the file was corrupted or modified — this is called a fixity check.

Tips for language documentation projects

  • Keep originals as backup. Before renaming or compressing, make sure you have a copy of your original files in a safe place.
  • One folder per session. Store all files from one recording session (video, audio, notes, subtitles) together in one folder. Then use the Organise tab to rename them all at once.
  • Use the right category. Choose the category that best describes the content — this helps future researchers understand what each file contains.
  • Use lowercase extensions. The "Use lowercase file type" option in the Organise tab ensures your files use .mp3, .mp4 etc. (not .MP3, .MP4). Lowercase extensions work better across different computers and operating systems.
  • Compress only for sharing. Compressed files lose some quality. Keep the full-quality original files for archiving and transcription; share compressed copies.
  • Language codes. Use the ISO 639-3 standard three-letter codes. You can look up any language at iso639-3.sil.org.

About OpenSpeaks Bento

OpenSpeaks Bento is part of the OpenSpeaks initiative, which builds open tools for language documentation and revitalisation. The tool is hosted on Wikimedia Toolforge and is free to use for everyone.

Changelog

1.0.0 — 2026-07-01

First stable release. Bento moves out of alpha after a series of redesigns and the addition of media analysis.

  • Adopted the Wikimedia Codex design system across the interface
  • Added the Analyse Media tab: format, codec, loudness, and source-device detection via MediaInfo.js, with SHA-256 fixity checksums
  • Accessibility fixes and inline recommendations throughout
  • Disabled caching for HTML/CSS/JS so users always get the latest version

0.0.2-alpha — 2026-05-27 to 2026-06-20

Several visual redesigns and a move to fully client-side compression.

  • Multiple full redesigns of typography, colour palette, and layout, including a dark mode
  • Moved audio and video compression entirely client-side using ffmpeg.wasm, removing the server-side dependency
  • Fixed contrast, mobile layout, and cross-origin isolation (COOP/COEP) issues

0.0.1-alpha — 2026-05-18

Initial alpha release.

  • Organise tab: drag-and-drop renaming with a configurable naming convention
  • Duration tab: per-file audio/video duration reporting
  • Compress tab: image, audio, and video compression

Development & Maintenance

People

Developer & maintainer: Subhashish Panigrahi

Significant improvements were made at the Indic Wikimedia Hackathon, Hyderabad 2026, based on input and feedback from Bharathesha Alasandemajalu, Jnanaranjan Sahu, and other participants at the event.

Contributions of code, translations, documentation, and bug reports are welcome from everyone.

Contributing code

The source code is hosted on GitLab (Wikimedia). Fork the repository, make your changes, and open a merge request. The frontend is a single self-contained HTML file with no build step required — you can open it directly in a browser to test changes.

Reporting bugs and suggesting features

Use the GitLab issue tracker to report problems or request new features. Please describe what you were trying to do, what happened, and which browser you were using.

Contributing translations

You can help translate OpenSpeaks Bento into your language — no programming knowledge required. Translations are managed through Translatewiki.net, a web-based platform used by Wikipedia and many Wikimedia projects:

1
Create a free account at translatewiki.net.
2
Search for OpenSpeaks Bento in the project list, or look under the OpenSpeaks group.
3
Choose your language and translate messages one by one using the web editor. No git or command-line knowledge needed.
4
Your translations are reviewed and merged automatically, and appear in the next release of the tool.
💡You can also contribute translations directly on GitLab. The locales/ folder contains one JSON file per language. Fill in missing strings and open a merge request.

Infrastructure

The tool runs on Wikimedia Toolforge. Audio and video processing — including compression — runs entirely in your browser using ffmpeg.wasm and MediaInfo.js. No files or data are sent to any server.