How Much Does Audio Restoration Cost?

The short answer on audio restoration cost: restoration is usually billed by the hour, because the time depends on the damage, and our Audio Restoration & Repair work starts from $65.00 per hour. When restoration is part of a finished product, the reviewed US market ranges for that product apply instead, such as $75 – $400 per podcast episode or $50 – $300 per finished minute of audio post for video.

That answer is accurate but not very useful on its own, because two files of the same length can need wildly different amounts of work. A clean interview with a little air-conditioner hum is a short job. A cassette recording of a family elder, captured on one microphone in a tiled kitchen with a refrigerator cycling in the background, clipped every time someone laughed, is a long one, and even after all that time the result has a ceiling. Understanding what sits between those two cases is what lets you budget sensibly and recognize a fair quote.

This guide is for podcasters, video producers, marketing teams, archivists, legal and corporate teams with recordings to rescue, and anyone who has been handed a file and told to "clean it up." It covers the pricing models suppliers use, how market ranges compare with our published starting rates, what drives the cost, an illustrative budget, how to get an accurate quote, the red flags in cheap quotes and how to budget for restoration over time.

Pricing models for audio restoration and what each includes

Suppliers package restoration in a few different ways. Each model shifts risk between you and the engineer, and knowing which one you are being offered tells you what questions to ask.

Hourly engineer time

Hourly billing is the most common model for pure restoration, because nobody can know in advance exactly how long a damaged recording will take. You pay for the engineer's time, usually with an estimate after an assessment and an agreement to check in before going over it. Our Audio Restoration & Repair rate starts from $65.00 per hour. Hourly work suits one-off rescues, archival material, legal or forensic-adjacent cleanup where you need a record of what was done, and anything where the damage is uneven across the file.

Per episode

Podcast producers usually buy restoration as part of an edited episode. The typical US market range for a podcast episode is $75 – $400 per episode, covering edit, cleanup, levels and music, mastered to a podcast target and priced per finished episode. Our podcast episode rate starts from $165 per episode and includes edit, cleanup, levels, music and a master at the right platform target, with a turnaround of 2–4 business days. Per-episode pricing works well when recordings are reasonably consistent. It breaks down when one episode arrives badly damaged, which is why most per-episode agreements define what "cleanup" covers and bill heavy repair separately.

Per finished minute

For video, audio post is often priced by the finished minute. The typical US market range is $50 – $300 per finished minute for dialogue editing, effects, music and a mix delivered against the video's loudness target. Our audio post for video starts from $95 per finished minute, covering dialogue, effects and music mixed against the picture and its loudness spec, with a turnaround of 3–5 business days. Dialogue restoration is part of that job, though very damaged location sound can push a project toward the top of any range.

Project fee

Commercial and corporate audio is often quoted as a project. The typical US market range is $400 – $2,500 for voice-over finishing, sound design and mixing to a broadcast or platform spec. Restoration inside a project fee is usually limited to what the supplier heard in the files at quote time, so send representative material before agreeing a price.

Dedicated engineer

Teams with steady volume sometimes want an engineer who works in their session format, follows their naming and delivers to their spec. Our dedicated engineer rate starts from $85 per hour, with a start within 48 hours. This model suits in-house teams that have their own editors but need specialist repair and mixing capacity on call.

Market ranges compared with our published starting rates

The table below sets the reviewed US market ranges beside our published starting rates. Where a market range is not available for a line, or where we do not publish a rate, the cell says so rather than filling the gap with a guess. Our figures are starting prices, not guaranteed totals; the final number depends on the material.

Two things are worth noticing. First, restoration is the only line priced purely on time, because it is the only one where the amount of work cannot be predicted from the length of the finished product. Second, the per-episode and per-minute ranges are wide for a reason: the bottom of each range assumes clean source material, and the top reflects heavy repair, complex sound design or many deliverables. For a deeper look at those adjacent services, see our guides on how much podcast editing costs and how much audio post-production costs.

What drives audio restoration cost

Every quote for restoration is really an estimate of engineer time. These are the factors that move it, roughly in order of how much they matter.

Runtime and raw material

The amount of raw material matters more than the length of the finished piece. Three hours of recording for a thirty-minute episode is an editorial job, not a cleanup job. Someone has to listen to all three hours, choose what stays and repair only what survives the edit. If you can make the editorial cuts yourself, or supply a paper edit with timecodes, you remove hours from the restoration bill because the engineer only repairs what will be heard.

Recording quality

Recording quality sets both the cost and the ceiling. One track per speaker in a quiet room is straightforward: each voice can be treated independently, noise can be gated between phrases and crosstalk is easy to manage. Four people on one microphone in a reverberant room is expensive, and the ceiling on the result is lower. When voices share a track, any processing applied to one affects all of them, and room reflections are baked into every word.

How much repair is needed

Noise reduction, de-reverb and clip repair are slow, and past a point they stop helping. Each type of damage has its own tools and its own limits:

• Broadband noise and hiss from tape, preamps or ventilation responds well to spectral noise reduction when it is steady, and poorly when it changes.
• Hum and buzz from electrical interference sits at the mains frequency and its harmonics; it is usually quick to remove when it is stable.
• Clicks, crackle and mouth noise can be removed automatically in light cases and need manual spectral editing in heavy ones.
• Clipping, where the waveform hit the ceiling of the recorder, can be reconstructed when it is brief and occasional; sustained clipping leaves permanent distortion. Our guide to repairing clipped audio explains where that line falls.
• Reverb from hard rooms can be reduced but not removed, and aggressive de-reverb makes voices sound thin and processed.
• Intermittent noises such as sirens, dogs, phones and chair squeaks often need manual work event by event, which is where hours accumulate.
• Tape problems such as dropouts, speed drift and wow and flutter on older formats need specialist tools and careful listening.

Engineers typically work in tools such as iZotope RX and CEDAR, which are powerful but not magic. Pushing any of them too far produces watery, warbling artifacts that are often more distracting than the original noise. A good engineer knows when to stop, and a fair estimate includes the time to listen back critically and undo processing that has gone too far.

Sound design

Music editing, effects and custom beds are creative work with their own hours. They are sometimes used to mask what cannot be fully repaired, for example a light music bed under a noisy interview. That can be an efficient choice, but it is additional work and should appear in the quote. Our piece on background audio and speech clarity covers how to do it without burying the voice.

Number of versions

Different loudness targets, stems and an M&E version for localization are each a deliverable. A podcast master, a broadcast master and a set of dialogue, music and effects stems are three outputs, not one. Common targets include roughly -16 LUFS integrated for podcasts, -23 LUFS under EBU R 128 for European broadcast and -24 LKFS under ATSC A/85 for US broadcast. Tell the supplier every destination up front.

Transcript accuracy

A raw automatic transcript is cheap. A corrected one with names and jargon right is a person's time. If the restored audio is going to be published with captions or a transcript, decide which you need; our guide to transcription and audio accessibility sets out the trade-offs.

How the use of the audio changes the job

The same damaged recording can cost different amounts to restore depending on where it is going. The destination decides how far the repair has to go, which spec it must meet and how many versions are needed.

Podcasts

Podcast listeners are forgiving of a little room character but not of fatigue. A voice that is slightly roomy but natural is easier to listen to for forty minutes than a voice that has been scrubbed until it warbles. The job is usually consistent levels between speakers, removal of distracting noises and a master at the platform target. Restoration budgets for podcasts are mostly spent on remote guests and on-location episodes. For the editorial side of that work, see podcast editing: the decisions that matter.

Video

For video, dialogue has to sit with picture. That means repair cannot shift timing, room tone has to match across cuts so edits do not jump, and the final mix has to meet the video's loudness spec. If the audio was recorded separately from the camera, sync has to be right before restoration starts, or repaired sections may need to be redone. When restoration is part of a full audio post job, the per-finished-minute model usually covers it, with heavy repair called out separately.

Archives

Archival restoration has a different goal. The priority is preserving the content faithfully, not making it sound modern. Good archival practice keeps an untouched preservation copy of the transfer and produces a separate access copy with restoration applied, so that future engineers with better tools can start again from the original. That second file is a deliverable, and documenting what processing was applied is part of the job.

Evidence and records

When recordings may be used in a legal, compliance or HR setting, the priority shifts again. Any processing needs to be documented, the original must be preserved unchanged, and enhancement should aim at intelligibility without altering what was said. That documentation takes time, and hourly billing is the natural fit. This is a practical note, not legal advice; the people handling the matter should set the requirements for how the audio is treated.

How a restoration project runs, and how timing affects cost

Time and cost are linked. A rush job compresses the listening and the checking, and it displaces other work, so it costs more. Here is how our restoration and related projects typically run.

The revision stage is where budgets drift. Vague notes such as "make it cleaner" lead to more rounds; specific notes such as "at 12:40 the fridge noise returns, and the host sounds muffled from 18:00 to 19:30" lead to fewer. Our guide to how long podcast editing takes goes further into scheduling.

A worked budget example (illustrative)

The example below is illustrative. The project, the file counts and the hours are invented to show how a budget comes together; the rates are our published starting rates, and the totals are starting figures, not a quote.

Imagine a small museum producing an oral history series. It has eight archival interviews on cassette that have already been digitized, one short documentary clip for its website and a plan to publish four of the interviews as podcast episodes.

The arithmetic: 12 × $65.00 = $780.00; 4 × $165 = $660.00; 6 × $95 = $570.00; and $780.00 + $660.00 + $570.00 = $2,010.00.

Three decisions in this example affect the real number. First, the twelve hours is an assumption; the assessment might find that two of the tapes are in far worse condition than the rest, or that most of them respond quickly to broadband noise reduction. Second, the museum's own staff could make the editorial selections from the long interviews, which would reduce the time needed to shape the podcast episodes. Third, if the museum later wants a broadcast master of the documentary clip as well as a web version, that is an additional deliverable.

The example also shows why mixing pricing models is normal. Pure restoration of archival material is billed on time because nobody can know how bad the tapes are until someone listens. The finished podcast episodes and the video clip have predictable outputs, so they fit per-episode and per-minute pricing.

How to get an accurate audio restoration quote

The quality of a quote depends almost entirely on what you send. A supplier who has heard your worst five minutes can price confidently; one who has only read a description is guessing, and guesses are either padded or wrong.

If you have a few files ready now, our audio editing and production services page explains how assessment works and what to send.

Reading a restoration estimate line by line

A well-built estimate tells you how the supplier plans to spend their time. Whether it is hourly or fixed, look for these components and ask about any that are missing.

• Assessment: whether listening to the material is included, and whether it is charged.
• Editorial work: selecting and cutting material, separate from repair.
• Repair: the types of damage identified and the estimated time for each, especially manual repair of intermittent noises.
• Mix and master: levels, tonal balance and loudness for each destination.
• Deliverables: the list of files, formats, loudness targets, stems and naming.
• Revisions: how many rounds are included and the cost of more.
• Change point: for hourly work, the point at which the engineer checks in before spending more time.

An estimate that breaks the work down this way is easier to compare with others and easier to control. If the repair line is larger than you expected, you can decide whether to cut the material more tightly, accept a lighter level of repair or re-record part of it. If deliverables push the number up, you can drop a version you do not really need. None of those choices is available when all you have is a single figure.

Red flags in cheap audio restoration quotes

Low prices are not a problem in themselves. Plenty of restoration jobs are genuinely small, and a clean file with steady hum can be fixed quickly. The problem is a low price that does not match what the material needs, because the difference gets made up somewhere: in skipped listening, in over-processing or in an invoice that grows later.

The most common failure in cheap restoration is over-processing. Automatic noise reduction pushed to its limit removes noise and, with it, the breath, sibilance and room tone that make a voice sound human. The result is quieter but harder to listen to for any length of time. A fair quote allows time for a person to listen through at the end and back off anything that went too far. Our guide to audio restoration done properly explains what a careful workflow looks like.

Questions to ask when two quotes differ widely

• Did both suppliers hear the same samples, including the worst section?
• Are both quoting the same deliverables and loudness targets?
• Does one include editorial work and the other only cleanup?
• How many revision rounds does each include?
• Will each deliver a short test before the full job?

How to reduce audio restoration cost before it starts

The cheapest restoration is the one you do not need. A few habits at recording time save more than any negotiation over rates.

• Record one track per speaker. Separate microphones and separate tracks make every later step faster.
• Choose the quietest, softest room available. Carpets, curtains and bookshelves reduce reflections; kitchens and glass-walled meeting rooms add them.
• Turn off what you can. Air conditioning, refrigerators, fans and notifications cause many of the noises engineers are paid to remove.
• Set levels conservatively. Leave headroom so laughter and emphasis do not clip; a slightly quiet recording is far easier to fix than a clipped one.
• Record a few seconds of room tone. Silence from the room gives noise reduction a clean profile to work from.
• Keep the originals. Never overwrite the raw files; our guide to archiving audio sessions covers a simple structure.

For material that already exists, the biggest saving is editorial. Mark the sections you actually want, supply timecodes and let the engineer spend their hours repairing only what will be heard.

How to budget for audio restoration over time

For one-off rescues, budget from an assessment and add a margin for surprises. For ongoing work, the picture is different, because the cost of restoration depends on habits that you can change.

Ongoing podcasts and video series

For a regular show, budget on the per-episode or per-minute rate and set aside a separate hourly allowance for occasional bad recordings, such as a remote guest on a poor connection. Track which episodes needed extra repair and why. If the same problem repeats, fixing it at the source, with a better microphone for a regular guest or a quieter room, usually costs less than paying to repair it every time.

Archives and back catalogs

Archives are best handled in batches. Assess a sample across the collection, group the files by type and severity of damage, and restore the most valuable or most at-risk material first. Batches let the engineer build processing chains that work for similar recordings, which is faster than treating every file from scratch.

Teams with steady volume

If restoration and mixing are regular needs, compare the per-item cost of your work with dedicated engineer time from $85 per hour, working in your formats and naming. The right choice depends on how predictable your volume is and whether you already have editors doing the first pass.

Plan for rights and reuse

Restored archival material often gets reused: in a documentary, a campaign or a training program. Before you spend on restoration, confirm that you have the rights to use the recordings the way you intend, especially if any AI-assisted tools are involved in the process; ask who owns the recordings, what the original participants agreed to and whether any tool used in the process changes those answers. Settling rights first avoids paying to restore material you cannot publish.

Is restoration worth it, or should you re-record?

Sometimes the honest answer to "how much will it cost to fix this?" is "more than re-recording." If the speaker is available, the content is scripted or easily repeated and the damage is severe, a new recording in a decent room will almost always sound better and cost less than hours of repair. Restoration earns its cost when the recording cannot be repeated: an interview with someone no longer available, a live event, a historical document, legal evidence or a moment that happened once.

A useful test is to listen to the worst minute and ask whether a listener would stay with it after it has been cleaned to a reasonable standard. If the answer is yes, restore it. If the answer is only "maybe," ask for a test clip before committing to the full job.