Auphonic

What Auphonic is

Auphonic is an automated audio post production service. It takes an audio or video file, processes it with a set of algorithms, and returns a cleaned, leveled, and loudness compliant result. The main audience is podcasters, but the same workflow fits educators, audiobook producers, video creators, and developers who need repeatable audio cleanup without manual mastering.

The product is browser based. You upload a file, choose a preset or individual algorithms, and start processing. The site also mentions external services such as SFTP, Dropbox, and Google Drive for slower uploads or larger workflows. That matters because Auphonic is not just a one off cleanup tool. It can sit between recording and publishing.

The core promise is consistency. Auphonic can reduce noise and reverb, balance speaker levels, control loudness, cut filler words, generate transcripts, and publish to platforms. Those tasks normally require separate plugins, manual editing, or an engineer. Auphonic bundles them into a credit based web service.

The free plan gives 2 hours of processed audio each month. That is enough to test the sound on a short episode, but free productions include an Auphonic jingle. Paid plans remove the jingle and add speech recognition, automatic shownotes, watch folders, batch productions, and higher monthly allowances.

Auphonic also points to adoption by more than 2M users and logos from BBC Radio, Buzzsprout, iHeartRadio, MSNBC, and ProSiebenSat1. Those logos do not prove fit for your show, but they show the service is aimed at professional audio environments, not just hobby recording.

Key capabilities

The central feature is the Intelligent Leveler. It balances levels between speakers, music, and speech. It applies dynamic range compression without requiring compressor knowledge. For interviews and panel recordings, this is often the difference between a usable episode and one where one host is too quiet and another peaks too hard.

Noise and reverb reduction

Auphonic can remove static noise, fast changing noise, and reverb. It can also keep or remove music. That choice matters. A podcast with theme music needs a different result from a lecture recording where only speech should remain. The service offers presets such as Voice Cleaner keep music, Voice Cleaner remove music, and Voice Cleaner remove breaths.

You can also adjust the amount of noise, reverb, or breath reduction. This avoids an all or nothing result. Heavy noise reduction can make voices sound thin. Being able to keep some room tone is useful when the recording is acceptable but not perfect.

Filtering, AutoEQ, and bandwidth extension

The filtering tools add missing frequencies and remove unwanted ones. AutoEQ creates separate time dependent EQ profiles for each speaker. It also includes De Esser and De Plosive handling. Bandwidth extension can recover high frequencies lost in low bitrate or archival recordings. This is useful for phone interviews, old tape, or compressed remote audio.

Automatic cutting

Auphonic can cut silent segments, pauses, coughs, throat clearing, sneezes, and filler words. The feature names examples like ah, uhm, mh, and ähm in multiple languages. You can run this without settings, or review cuts manually in the Audio Inspector. You can also export cut lists for fine tuning in another editor.

This is not the same as a full human edit. It will not judge story pacing or decide whether a pause is intentional. It removes obvious dead air and mouth noise. For many podcasts, that is enough. For narrative audio, you may still want manual review.

Multitrack processing

Multitrack productions process multiple input files into one mixdown. Auphonic handles speech tracks from multiple microphones, music tracks, and remote speakers. It applies automatic ducking, adaptive noise gates, and mic bleed removal. Billing is based on the duration of the output file, not the sum of all input tracks.

The free plan limits multitrack productions to under 20 minutes. Paid plans remove that limit. If your show has several hosts and separate track recordings, this is one of the main reasons to pay.

Loudness and output specs

Auphonic can target loudness specifications such as -16 LUFS for podcasts. It also supports true peak limit, MaxLRA, and related parameters. It lists targets for Netflix, Audible, Spotify, podcasts, EBU R128, ATSC A/85, radio, and commercials. This is useful if you deliver to platforms with strict audio requirements.

Speech to text and shownotes

Paying users get multilingual speech recognition. Auphonic uses a self hosted Whisper model and also supports external speech recognition services such as Amazon Transcribe, Google Cloud Speech API, wit.ai, and Speechmatics. The transcript editor is shareable. In multitrack productions, each track can be processed separately so the transcript shows which speaker is active.

Automatic shownotes and chapters generate summaries and timestamped sections. This can save time on episode pages, but the value depends on your content and tolerance for automated summaries. You should still review the text before publishing.

Video, metadata, and publishing

Auphonic supports video input. It extracts audio, processes it, and merges it back with the original video track without loss of image quality. It can create audiograms with waveform visuals, chapters, and metadata. It can also publish results to platforms such as YouTube, Libsyn, PodBean, Soundcloud, and Facebook.

Recent updates add burn in subtitles, automatic video cutting, a command line interface, and direct publishing to RSS.com, Acast, and Scrybecast. These are workflow features. They matter if you produce often and want fewer manual steps between recording and publication.

Auphonic also exposes an API and CLI. The homepage shows a production API example with a preset, input file, and start action. Watch folders and Zapier are listed for automated workflows. These options make sense for teams or products that process audio in volume, not for a single casual recording.

Pricing explained

Auphonic bills by duration of processed audio. The minimum charge is 3 minutes per production. If you process a 20 minute file with a 1 minute intro, the processing time is 21 minutes. Multiple output formats do not increase the charge. Multitrack billing is based on the output duration, not the total of all input tracks.

The free plan includes 2 hours each month. Free credits do not roll over. Free productions include an Auphonic jingle. The free plan also excludes speech recognition, automatic shownotes, watch folders, and batch productions. It limits multitrack to under 20 minutes.

Paid recurring credits come in S, M, L, XL, and XXL sizes. On the yearly view, S is $11 per month with $132 billed yearly and 9 hours per month. M is $25 per month for 21 hours. L is $52 per month for 45 hours. XL is $113 per month for 100 hours. XXL is $245 per month for 250 hours. The monthly view is higher: S is $13, M is $30, L is $64, XL is $136, and XXL is $290.

Recurring credits reset every month. Unused recurring credits do not stack. That is the main trade off against one time credits. One time credits never expire. They are used after recurring credits but before free credits. You can enable auto renew so a one time package tops up your balance when it falls below 1 to 10 hours.

One time credit pricing starts at $12 for 5 hours, or $2.4 per hour. It drops to $2.3 per hour for 10 hours, $2.2 for 25 hours, $2.0 for 50 hours, and $1.71 for 100 hours. Larger packages go lower: $1.28 per hour for 250 hours, $1.09 for 500 hours, $1.045 for 1000 hours, $1.01 for 2000 hours, and $0.975 for 3000 hours.

You can combine recurring and one time credits. This is useful for a regular monthly show with occasional heavy months. If you buy both, recurring credits are used first. If you edit and rerun a production with the same input files, you are not charged again. If you change input files or create a new production, you are charged again.

Auphonic says credits are charged only when processing succeeds. It also offers a refund credits link if you describe a problem with a result. Those terms reduce the risk of paying for failed or unusable output, but you still need to check whether the sound fits your voice and format before buying a large package.

Prices are listed without VAT. Sales tax or VAT may apply depending on where you live. European companies can use a VAT ID to purchase without VAT. That is a small detail, but it matters for business buyers who need clean invoices.

How it compares

Auphonic is not a full digital audio workstation. It does not replace manual mixing when you need surgical edits, creative sound design, or detailed control over every transition. It is closer to an automated mastering and cleanup layer. You give it source audio, and it returns a publishable file.

Compared with doing nothing, the value is obvious. A raw recording often has uneven levels, room noise, plosives, and dead air. Auphonic addresses those problems in one process. Compared with manual editing, the value is time. You lose some judgment, but you gain speed and consistency.

Compared with hiring an audio engineer, Auphonic is cheaper for routine work. A 21 hour monthly recurring plan costs $25 per month when billed yearly. A 5 hour one time package costs $12. That makes it easy to test on real audio without a large commitment. The downside is that automated processing may not match a skilled engineer on difficult recordings.

Compared with a local plugin suite, Auphonic is less hands on. Plugins give immediate visual feedback and unlimited renders on your machine. Auphonic uses credits and requires upload. In exchange, it offers loudness targets, transcripts, publishing, and API access that most plugin bundles do not provide as one pipeline.

The free plan is a real trial, but it is not a production plan for a public show. The jingle makes that clear. If you publish regularly, the paid tiers are the practical product. The yearly recurring plans are cheaper per hour, while one time credits fit irregular volume.

Who should skip it

Skip Auphonic if you need frame accurate manual editing for narrative audio. It can cut filler words and silence, but it is not a replacement for a human editor who shapes story timing. It also does not replace a DAW for music production or complex sound design.

Skip it if you only need occasional basic cleanup and already have tools that do that for free. If your recorder produces clean audio and your editor handles loudness, Auphonic may be redundant. The strongest case is repeated production, not one off repair.

Skip it if speech recognition is your only need and you do not want to pay for a recurring plan. The free tier does not include speech recognition. A dedicated transcription service may fit that job better if you do not also need leveling, noise reduction, and loudness control.

Skip it if unused monthly credits bother you. Recurring credits reset every month. If your publishing schedule is unpredictable, one time credits are safer, but they cost more per hour at small quantities. The pricing model is flexible, but you need to match it to your volume.

Also skip it if you need a desktop offline tool. Auphonic is a web service with external file transfer options, not a local application you run without an internet connection. If your studio policy restricts cloud processing, that may be a deal breaker.

Consider it if you produce speech audio on a schedule and want a predictable output standard. The best fit is a podcast, lecture series, audiobook, or video channel where the same cleanup and loudness target should apply to every file. Start with the free 2 hours, then buy the smallest one time package or recurring plan only after the output works for your voice.