Auphonic

New

Web service that levels speech and music, normalizes loudness to set targets, reduces noise and reverb, cuts filler words and silence, and adds transcripts, chapters and publishing for podcasts.

auphonic.com

Auphonic

Explore features, practical uses and pricing below.

Auphonic is a web service for post-producing spoken audio. You upload a recording, pick which processing to apply, and get back a file that's leveled, cleaned and normalized to a loudness target, with optional transcripts, chapters and publishing to hosting platforms. It's run by Auphonic GmbH. There's no timeline editor; Auphonic processes files and leaves content editing to other tools.

The processing chain, step by step

  • Adaptive Leveler balances speakers, music and speech, applying dynamic range compression so a quiet guest and a loud host end up at comparable levels.
  • Loudness Specifications set integrated loudness, true peak, MaxLRA and other targets. The site gives -16 LUFS for podcasts as an example and mentions EBU R128 and ATSC A/85.
  • Noise & Reverb Reduction can remove only static noise or also fast-changing noise, and can keep or remove music. Noise, reverb and breath reduction are adjusted separately.
  • Filtering & Enhancement includes AutoEQ with de-essing and de-plosive processing, a Studio Voice model, repair of clipping and codec artifacts, and bandwidth extension for dull recordings.
  • Automatic Cutting removes silences, coughs and filler words such as "ah" and "uhm" in multiple languages. Cuts can be edited in the Auphonic Audio Inspector or exported as cut lists.

Multitrack, transcripts and automated publishing

For interviews recorded on separate microphones, the multitrack algorithms process each track and then mix them with adaptive noise gates, mic bleed removal and automatic ducking of music beds. Sync Tracks aligns files recorded on different devices. Speech2Text runs a self-hosted Whisper model or external engines such as Amazon Transcribe, Google Cloud Speech and Speechmatics, and can produce show notes and chapters. Finished files can be sent to YouTube, Libsyn, PodBean, SoundCloud and other services, and the Auphonic API, CLI, watch folders and Zapier support automated pipelines.

Producers who fit Auphonic best

It suits podcast producers who want every weekly episode at the same loudness, radio and audiobook teams with delivery specs, and developers adding audio processing to a product through the API or the white-label option. It's a poor fit if you want to restructure an interview by cutting and moving sections. Pair it with a DAW or a transcript editor such as Descript for that, and browse other AI podcast tools if you need recording as well.

A two-person interview through Auphonic

Record host and guest on separate tracks, create a multitrack production and upload both files with an intro music file. Turn on the Adaptive Leveler, noise reduction with music kept, and automatic ducking so the intro drops under speech. Set the loudness target to -16 LUFS and enable Automatic Cutting for fillers, then check the cuts in the Audio Inspector. Add Speech2Text for a transcript and chapters, and publish to your podcast host.

Where automatic processing needs a listen

Strong noise reduction can thin out a voice, and removing fast-changing noise may affect laughter or crosstalk, so check loud and quiet passages before publishing. Automatic cutting can remove a deliberate pause, which is why the cut list matters. Billing counts file duration, with a three-minute minimum per production. Free productions carry an Auphonic jingle, multitrack is limited to under 20 minutes, and multilingual speech recognition and automatic show notes aren't included on the free tier.

Credits and plans

The pricing page says Auphonic is free for two hours of processed audio a month, and unused free credits don't carry over. Paid options are recurring monthly credits from Auphonic S at 9 hours up to XXL at 250 hours, larger custom plans, or one-time credits that never expire. Running a production again with the same inputs and new settings isn't charged, and failed productions aren't billed.

Auphonic questions

Does Auphonic edit video?

It processes the audio of video files and can add subtitles, chapters and waveform audiograms, but it isn't a video editor.

Which transcription engine does it use?

Auphonic runs its own Whisper model and can also connect to Amazon Transcribe, Google Cloud Speech, wit.ai or Speechmatics if you already use one of them.

Similar tools

Record podcasts and video interviews with local multitrack capture up to 4K, then edit by transcript, clean audio with Magic Audio, cut short clips, translate, livestream and publish from one platform.
Browser tool that removes filler words, stutters, mouth sounds, long silences and background noise from podcast audio, with multitrack sync, timeline export to DAWs, transcripts and an API.
Turn supplied music or audio into waveform videos with custom artwork, text, visual effects and automated lyric visualization.