> For the full page index, fetch https://echonova.ai/llms.txt # Sound Cleaning formats, limits and credits The formats the Sound Cleaning page accepts, every limit with the message it shows, what a cleanup costs in credits and when the credits are taken. This page is the lookup table for the **Sound Cleaning** page: the formats it takes and the one it gives back, every limit with the message you see when a file crosses it, the price of a cleanup in credits and the moment it is charged, and what the page shows when the workspace balance cannot cover a file. The steps are on [Clean a recording](/docs/playground/clean-a-recording). ## Formats | Direction | Formats | | ------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------ | | Input, by file extension | MP3, WAV, FLAC, OGG/Opus (`.ogg` or `.opus`) | | Input, by contents | The file must contain audio. A file with the right extension but other contents is refused after upload. | | Output | WAV, always. The download is named `cleaned_` followed by your original file name without its extension, and `.wav`, so `call.mp3` is saved as `cleaned_call.wav`. | M4A, AAC and WEBM are accepted by the Speech to Text page and not here; export such a recording as MP3 or WAV first. The **Cleaned** card shows the cleaned file's loudness in LUFS; the peak level, sample rate and bit depth are not shown, so read them from the file in your audio tool. ## Cleaning settings Four presets in a row under the drop area, and eight values in the **Advanced** panel, grouped as **Low cut**, **Loudness** and **Dynamics**. A preset is a set of values; **Studio Master** sends none and leaves every value at its default. | Preset | Values it sets | | ------------------- | -------------------------------------------------------------------------------------------------- | | **Studio Master** | Nothing; every default below | | **Heavy AC & Wind** | Rumble & hum cutoff 120 Hz, Target volume -16 LUFS, Maximum peak level -1 dBFS | | **Radio voice** | Rumble & hum cutoff 90 Hz, Loud spikes trigger -6 dBFS, Volume smoothing 5, Target volume -14 LUFS | | **Deep Voice** | Rumble & hum cutoff 70 Hz, Target volume -16 LUFS | | Slider | Default | Range | What it does | | ----------------------- | --------- | -------------- | -------------------------------------------------------------------------------------------------- | | **Rumble & hum cutoff** | 90 Hz | 20 to 300 Hz | Cuts low sounds: air conditioning, mic bumps, hum. Raise for wind or rumble; lower for deep voices | | **Filter smoothness** | 0.707 | 0.1 to 5 | How gradual that low cut is. Rarely needs a change | | **Target volume** | -16 LUFS | -30 to -6 LUFS | Overall loudness. -16 for podcasts and video, -14 for music platforms, -18 for background speech | | **Maximum peak level** | -1 dBFS | -6 to 0 dBFS | The loudest the file may reach. Keeps playback free of crackle | | **Loud spikes trigger** | -3.8 dBFS | -30 to 0 dBFS | Where loud peaks start being tamed. Lower for wild swings, raise for dynamic speech | | **Volume smoothing** | 4 | 1 to 20 | How strongly peaks are reined in. 2 gentle, 4 broadcast, 6 aggressive | | **Reaction speed** | 5 ms | 0.1 to 100 ms | How fast a loud syllable is tamed | | **Recovery speed** | 50 ms | 1 to 1000 ms | How fast volume returns to normal | A value outside its range is refused before upload. The values used are kept with the cleanup, and its preset name shows on the **Cleaned** card (**Custom** when the values match no preset). ## Limits | Limit | Value | Applies to | Message when crossed | | --------------- | ------------------------------------- | ----------------------------------------------------------- | -------------------------------------------------------------------------------- | | Audio length | 1 hour | Every file, checked in the browser before upload | **That clip is longer than 1 hour. Split it and try again.** | | Audio length | 1 hour | Every file, checked again after upload on the decoded audio | **Audio exceeds the 1-hour limit.** | | File size | 150 MB | Every file, checked before upload | **That file is over 150 MB. Trim it or export a smaller version.** | | File size | 150 MB | Every file, checked again after upload | **Audio file exceeds the 150MB limit.** or **File too large** | | Format | MP3, WAV, FLAC, OGG/Opus | Every file, by extension, before upload | **That file type is not supported. Use MP3, WAV, FLAC or OGG/Opus.** | | File type | Must contain audio | Every file, after upload | **Unsupported file type: … Please upload an audio file.** | | Format match | The file's type matches its extension | Every file, after upload | **Unsupported audio format. Sound cleaning accepts MP3, WAV, FLAC or OGG/Opus.** | | Readable file | The browser must be able to decode it | Every file, before upload | **We could not read that file** | | Readable length | The server must be able to decode it | Every file, after upload | **Could not read the audio length. Try another file.** | | Files at a time | One | The drop area takes one file per drop | No message | | Cleaning time | 10 minutes | Every run, from the moment the file is accepted | **Sound cleaning is temporarily unavailable. Please try again.** | | Cleaned files | The newest 8 | The list on the Sound Cleaning page | No message | Every message appears in the status line under the box, in red. After a check before upload, nothing was sent, so drop another file. After a message that arrived after upload, a **Try again** button appears beside **Clean** and sends the same file again, and the message is also shown as a notice titled **Error**. Everyone in the workspace can use the page. ## Credits | Fact | Value | | ------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | | Rate | 1,000 credits per minute of audio, charged per second. | | What is measured | The length of the audio as the server decodes it, not as the file's header claims, including any fraction of a second. A file whose length cannot be measured is refused with **Could not read the audio length. Try another file.** | | Rounding | Per second, with a minimum of 1 second. The charge is rounded to a whole credit, at least 1. | | When it is taken | After the file is received and passes the checks after upload, before cleaning starts. | | Refunds | The whole charge is refunded if the run stops for any reason, including the 10-minute limit. A run that stops leaves nothing in **Cleaned files**. | | Adjustment | None. The charge is computed from the length before cleaning and does not change when the cleaned file arrives. | | Replay and download | Free. Playing or downloading a cleanup any number of times is not charged. | | Where you see it | The credit counter in the sidebar, shown as used out of total with the label **Used**, refreshes when the cleaned file arrives. **Credits remaining** in the user menu shows the same balance. The page itself shows no price. | The rate is set by hand and can change; a cleanup is priced at the rate in force when it runs. The rates for every service are on [Credits](/docs/developers/credits). ### Worked examples At today's rates: | Audio | Charged | | ---------------------------------- | -------------- | | A 30-second voice note | 500 credits | | A 5-minute call | 5,000 credits | | A 12-minute interview | 12,000 credits | | A 1-hour file, the longest allowed | 60,000 credits | Every second costs the same; nothing is rounded up to a minute. ## When the balance cannot cover a file The charge is checked against the workspace balance before cleaning starts. If the balance is short, nothing is cleaned and nothing is charged. You see the message twice: as a notice titled **Error**, and in the status line under the box with a **Try again** button. | Message | Meaning | What to do | | ------------------------------------- | --------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------- | | **Insufficient subscription balance** | Your credit balance is less than the charge for this file. | Open **Subscription & billing** from the user menu and add credits, or clean a shorter file. | | **Subscription has expired** | No paid plan and no extra credits left. Basic limits apply, with no plan credits. | Choose a plan or add credits under **Subscription & billing**. | **Try again** sends the same file again. Press it only after the balance has changed; the check runs again and fails the same way otherwise. ## FAQ ### What does a short clip cost? Its seconds at 1,000 credits a minute: a 3-second clip costs 50 credits. ### Is a refused file charged? No. Every check before and after upload runs before the charge is taken. A file refused in the browser never leaves your computer. ### Is a run that stops charged? No. The charge is taken when the file is accepted and refunded when the run stops, including when it passes the 10-minute limit. ### Where can I see what one cleanup cost? The page does not show a per-file price. Work it out from the length at 1,000 credits per minute, or watch the sidebar counter before and after. ### Can I get MP3 back instead of WAV? No. Every cleaned file is WAV. Download it and convert it with a tool of your choice; the conversion costs nothing. ### Can I raise the 1-hour limit? No. Split the recording and clean each part. Each part is charged by its own length, so the total is the same. ## Related - [Clean a recording](/docs/playground/clean-a-recording): the steps, and where each message appears. - [Speech to Text limits and credits](/docs/playground/stt-limits-and-credits): the same table for transcription, which accepts longer files. - [Speech to Text formats](/docs/playground/stt-formats): what the browser does to each format before a transcription. - [Sound Cleaning quickstart](/docs/developers/clean-quickstart): the same cleanup from your own code, up to 10 minutes per file. - [Credits](/docs/developers/credits): what a credit is, rates for every service, and how a key's credit limit works over the API.