Audio is extraordinarily important to many blind people. It may be how we preserve a family memory, publish a podcast, teach a lesson, record an interview, produce a spoken-word programme, share music or capture an event we want to experience again.
Making the recording is only the beginning.
One speaker may be considerably quieter than another. A fan may hum in the background. Traffic may intrude through an open window. A room may sound hollow and echoey. A sudden word may be uncomfortably loud, or a voice may sound dull, boomy or harsh. Music intended to sit quietly behind speech may overwhelm it instead. Even when every word is present, the listener can be left constantly adjusting the volume.
Auphonic is an online service designed to help with problems such as these. It can balance quiet and loud voices, reduce distracting noise, lessen room echo, control sudden loud moments, improve aspects of voice tone, bring programmes towards a consistent listening level and combine separate tracks into a finished production.
The results can be extremely impressive. But for a screen reader user, reaching those results through a web interface can involve a lot of navigation, particularly when a production contains several tracks or when settings need to be adjusted and compared.
That is why Accessible Auphonic has been created.
Accessible Auphonic is a new Windows application produced by Hartgen Consultancy in association with Desiree Simione. It provides an accessible, keyboard-driven way to choose recordings, select the processing you want, send the material to Auphonic and download the completed files.
The application will be available from the Hartgen Consultancy website within the next few days. It will be completely free of charge.
And, unusually for a Hartgen Consultancy application, it has not been designed solely for JAWS. Accessible Auphonic works with both JAWS for Windows and NVDA, and it has been thoroughly tested with the latest version of NVDA.
The Personal Story Behind Accessible Auphonic.
There is a very personal story behind this application.
Desiree Simione has loved audio since childhood. That interest became even more important when she became totally blind at the age of 13. She would walk around her home, neighbourhood and school recording anything she could, sometimes being told by adults to put the tape recorder away and concentrate on something else.
For Desiree, audio performs something of the role that photographs perform for many sighted people. A recording can preserve not only what someone said, but the complete sound surrounding a memory: voices, movement, atmosphere and a sense of being back in that place. Recordings of her three daughters are especially valuable to her for that reason.
But recording sound was never enough. Desiree wanted to edit it, shape it and turn it into finished productions. Before accessible audio tools were readily available, she used two tape recorders to imitate multitrack recording, spending hours combining speech, sound effects and background music. Later, a Tascam four-track recorder gave her more freedom to experiment.
As technology advanced, she explored many audio editors. Although accessibility improved, the experience could still be frustrating. Sometimes additional scripts or workarounds had to be purchased merely to gain access to part of a mainstream product.
About a year ago, Desiree began using Auphonic to make her recordings cleaner and more professional. She liked what it could do, but found that its largely accessible web interface could still be cumbersome to navigate, particularly for multitrack productions. She wanted a faster and more comfortable way to use the service.
She had also begun exploring vibe coding: using artificial intelligence to help turn an idea into a working application. This was a new skill, but she had already created several tools to make web-based tasks easier, including an application for software used in her work. She decided to see whether the same approach could provide a more accessible desktop interface for Auphonic.
Desiree tests everything she builds by attempting to complete real tasks herself. When something is unclear, inaccessible or more awkward than it should be, she continues refining it. That practical approach is woven throughout Accessible Auphonic.
She invited Brian Hartgen to try the application after seeing his collaboration with her husband, Jonathan Simeone, on PresentAble. Brian was immediately interested, suggested ways in which the experience could be strengthened, and began writing the extensive documentation which accompanies it. The collaboration has reached the point where Brian now uses Accessible Auphonic for his own audio work.
Desiree also brings considerable professional accessibility knowledge to the project. For the past nine years, she has worked for Pearson Education as a Quality Assurance Accessibility Specialist, testing examinations and certifications with JAWS, NVDA, Narrator and VoiceOver. Away from work, she sings, records music and produces the Learning Blind Tech with Desiree Renae podcast. Her songs and podcast can be found at:
This is therefore an application created by someone who cares deeply about sound, regularly works with accessibility, uses both JAWS and NVDA, and understands the difference between a feature being technically reachable and a task being genuinely comfortable to complete.
What Does Auphonic Actually Do?
If you have not worked with audio processing before, some of the terminology can sound intimidating. The central idea is much simpler than the names suggest.
You begin with a recording which already exists on your computer. Accessible Auphonic lets you choose the improvements you want and sends a copy of the recording to the Auphonic online service. Auphonic processes it, and Accessible Auphonic downloads the finished version to a folder on your computer.
Your original recording remains where it was. It is not overwritten. This means you can compare the original and processed versions, try a different collection of settings or return to the untouched source later.
Accessible Auphonic is not a conventional audio editor. It does not record the original material, remove a mistake from the middle of a file, rearrange clips or move one multitrack recording into time with another. Those jobs should be completed in a recording or editing application first. Accessible Auphonic begins when you have existing audio which you want Auphonic to improve, balance, combine or prepare for distribution.
Four Clear Working Areas.
The main window is divided into four areas: Singletrack, Batch, Multitrack and History. Each can be reached with a simple shortcut or by moving through the standard Windows controls.
Singletrack is used for one existing audio or video file. That file might contain one speaker, an interview which has already been mixed, a lecture, a song, a podcast episode or a complete programme containing speech and music. A stereo recording is still one file and belongs in Singletrack.
Batch is for several files which need similar treatment. Perhaps you have recorded a series of lessons, a collection of spoken articles or several podcast episodes under similar conditions. You choose the files and settings, and each file becomes its own separate Auphonic production. Batch does not join them together; it saves you from repeating the same setup for each recording.
Multitrack combines separate recordings into one finished programme. You might have one microphone file for the interviewer, another for the guest and a third containing background music. Each track can be given its own role and treatment before Auphonic creates the final mix. The tracks must already line up correctly in time, but Accessible Auphonic gives you a structured way to describe how those prepared tracks should relate to one another.
History connects to your Auphonic account and lists earlier productions, including work started outside Accessible Auphonic. You can inspect the available output files and download the one you need. This is particularly helpful if an earlier automatic download was interrupted or if you want another format created by the same production.
Your First Finished Recording.
The simplest way to begin is with a short recording you know well. Listen to the original and identify one specific problem. Perhaps a guest is quieter than the presenter, or a low background hum makes the words tiring to follow.
In Accessible Auphonic, select Singletrack, choose the recording and check the production title. Open the processing settings and select the treatment which addresses the problem. Then choose the output format, such as an MP3 listening copy.
When you start the production, Accessible Auphonic does not silently upload the file. It displays a confirmation explaining that the recording will be sent to Auphonic and may use some of your processing allowance. You decide whether to continue.
The application then reports its progress while the file is uploaded, processed and downloaded. When the work is complete, you can open the outputs folder and listen in your usual media player. Status reports can be reopened and read if a spoken message passed too quickly.
The most important part is the final comparison. Are the words easier to hear? Do the voices sound natural? Has wanted music or atmosphere been preserved? Has the particular problem you identified actually improved? Audio processing is not about switching on every available option. It is about listening, making a considered choice and judging the result.
Reducing Background Noise Without Losing What Matters.
The guide explains several denoising methods in practical terms because not every kind of noise should be treated in the same way.
The Dynamic denoiser is intended for distracting sounds which change during speech while retaining voice and music. Speech isolation is more forceful and is useful when the voice is all that matters; it may remove music and the natural atmosphere of the location. Static denoising can work without needing a quiet section and aims to preserve music and ambience. Classic denoising automatically uses quiet portions of the recording and includes options for dealing with electrical hum.
There are also settings for reducing room echo and audible breaths. The fact that an option exists does not mean it should always be used. If a recording begins to sound watery, muffled or artificial, the answer may be a gentler setting. If the atmosphere of a location is important to the story, removing all of it may make the result technically cleaner but emotionally poorer.
The documentation repeatedly encourages careful listening rather than the pursuit of the strongest possible processing. That philosophy is important. Auphonic can improve difficult recordings, but it cannot recreate detail which was never captured or guarantee that every aggressive treatment will sound natural.
Making Voices Comfortable to Hear.
Different voice problems call for different tools.
The Adaptive leveler can reduce the difference between quiet and loud passages so that the listener does not need to keep reaching for the volume control. Compression can control brief, sudden loud moments. Loudness normalization can bring the complete programme towards a consistent target, which is useful when publishing a series whose episodes should play at a similar overall level.
Terms such as LUFS, RMS, compression and target loudness can appear highly technical when encountered in a list with no explanation. Accessible Auphonic provides context help for individual controls, while the User Guide explains what the terms mean, why a choice might be useful and what you should listen for in the result.
Voice tone can also be adjusted. Auphonic includes treatments which may help a voice which sounds dull, boomy or harsh. More advanced options can attempt to restore high-frequency character to restricted recordings or reconstruct damaged speech more extensively. These can be powerful, but they can also change how a familiar person sounds. The guide therefore recommends making an audition and listening to recognisable phrases before applying strong treatment to important material.
Audition Before Processing the Complete Programme.
One of the most useful features is Audition.
When suitable uncompressed WAV source files are used, Accessible Auphonic can prepare temporary excerpts and ask Auphonic to process a short sample using your current choices. You can then play the result from within the audition window.
If it does not sound right, return to the settings, change one choice and try again. If it does, you can start the full production using those settings.
This gives you a disciplined way to experiment. You can hear whether noise removal is too strong, whether voices have become easier to follow, or whether wanted music has been damaged before committing to the complete recording. Each audition is a separate Auphonic job and uses processing credit, so the app and guide make that clear as well.
Presets Without Losing Flexibility.
A preset is a saved collection of Auphonic settings. You might create one for a weekly interview, another for spoken lessons recorded in a quiet room and another for a particular podcast format.
Accessible Auphonic can load presets from your Auphonic account. You can save a new preset, rename one, update it or remove it. A useful preset can dramatically shorten a repeated workflow.
But presets do not lock you in. A custom choice made in Accessible Auphonic can replace the corresponding part of the preset for the current production while leaving its other settings intact. For example, you could use a familiar interview preset but turn denoising off for one episode because the background atmosphere is important.
The application helps you review these interactions, and the guide explains them carefully. A preset name alone does not prove that it suits the current recording. You remain responsible for listening and deciding.
Choose the File Your Listener Needs.
Accessible Auphonic supports a wide range of output formats, including MP3, AAC in M4A or MP4, 16-bit and 24-bit WAV, FLAC and Opus. You can also use the output formats stored in an Auphonic preset.
The guide explains the practical difference between an output format, bitrate and sampling rate. It discusses the balance between file size and quality and reminds you that a publishing service may have particular requirements.
This matters because the best choice depends on what happens next. A compact MP3 may be ideal for routine listening or podcast distribution. WAV may be preferable when the file will undergo further editing. FLAC provides lossless compression. The aim is not to memorise technical definitions but to understand enough to make a sensible choice.
Process Several Recordings Efficiently.
Batch processing is designed for repetition without unnecessary effort.
Select several recordings, review the list, choose the processing and output settings, and start the batch. Accessible Auphonic sends the files one at a time and provides a final report listing successes, saved locations and any failures. If one file fails, the remaining recordings can still be attempted.
A separate Batch audition is not needed. Instead, the guide recommends trying a representative file in Singletrack first. Once you are satisfied that the settings suit that material, apply them to the batch. This is a good example of the application and documentation working together to encourage an efficient but cautious workflow.
Bring Separate Tracks Together.
Multitrack production is where a web interface can become especially demanding, and it is one of the areas in which Accessible Auphonic has the greatest potential to help.
After adding the prepared tracks, review them in a list and select one or several for adjustment. A track can be identified as foreground speech, background sound or music whose volume should be reduced while someone is speaking. This last process is often called ducking. Gain can raise or lower a track by a fixed amount, and each track can receive appropriate processing before the final mix is created.
Imagine an interview with separate microphone recordings for two speakers and a music track. One microphone may need noise reduction while the other does not. The voices should be prominent. The music should support the programme without obscuring the words, becoming quieter beneath speech and returning naturally afterwards. Accessible Auphonic gives the screen reader user a structured route through those choices.
Once the individual tracks are ready, choose the loudness and output settings for the complete mix. You can audition suitable WAV tracks or process the full programme. The app does not pretend that multitrack production requires no judgement, but it makes the available decisions far easier to locate, understand and control.
Accessibility Which Extends Beyond Tabbing.
Accessible Auphonic uses standard Windows controls, allowing familiar navigation with Tab, Shift+Tab, Arrow keys, Enter and Space. Each main working area has a direct shortcut, and common actions such as browsing for a file, opening processing settings or starting a production can also be reached quickly.
There is a searchable Command Centre. If you know the action you want but cannot remember its location or shortcut, open the Command Centre, type part of its name and choose it from the results. Commands from another page can take you to that page before opening the correct settings.
Pressing Shift+F1 on a control opens detailed help for that control and, where relevant, its current choice. For a sound treatment, the help explains what it is intended to do and what you should listen for. F1 opens the complete User Guide.
Reports are placed in read-only Edit Fields so they can be explored with normal screen reader commands and copied when necessary. Progress and status information are exposed to supported screen readers, with speech and Braille feedback where available. The application remembers useful locations such as the last source folder and the folder for completed recordings, while leaving you in control of the files selected for each new session.
JAWS users will find familiar conventions in the application, but NVDA support is not an afterthought. The latest version of NVDA has been used extensively during testing. This makes Accessible Auphonic an important departure for Hartgen Consultancy and opens the app to a much wider group of screen reader users.
A User Guide Which Teaches the Subject.
The Accessible Auphonic User Guide is unusually extensive for a free application.
It does more than identify controls and list keystrokes. It begins with a first recording, explains why you might choose a particular treatment, suggests what to listen for, and gradually introduces more advanced work.
There are complete chapters on connecting an Auphonic account, keyboard navigation, presets, noise reduction, voice leveling, voice tone, output formats, auditions, batch processing, multitrack production, history and troubleshooting. A full keyboard reference is included.
The final chapter is a plain-language glossary of audio terms. It explains concepts such as bitrate, compression, decibels, denoising, ducking, gain, leveling, loudness normalization, LUFS, RMS, presets, productions, reverb and tracks.
This matters because many people have been made to feel that serious audio work belongs only to those who already speak the language of sound engineering. The guide does not assume that knowledge. It gives people a route into the subject, explains each idea when it becomes useful and encourages them to trust careful listening rather than simply selecting the most powerful-sounding option.
Install It, or Carry It with You.
Accessible Auphonic requires 64-bit Windows 10 or Windows 11 and will be available in two forms.
The installable edition adds the program to Windows, can create Start Menu and desktop shortcuts, and includes straightforward options for updating, repairing or removing it.
The portable edition runs from a folder you choose and does not need to be installed. It can be kept in a writable folder or carried on a USB drive. Its application files, settings and recordings remain together, although an encrypted Auphonic API key may need to be entered again when the portable copy is moved to another computer.
Both editions can check for updates from within the application. Update details can be read before anything is installed, and the app checks the completed download before replacing the existing version.
The Account and Processing-Cost Distinction.
Accessible Auphonic is free. There is no purchase price, subscription or activation fee for the Windows application.
Auphonic itself is a separate online service. You need an Auphonic account and an API key so that the application can send productions with your permission. At the time the guide was completed, Auphonic offered a monthly free processing allowance, with optional paid plans and one-time credits for people who need more. Free productions were subject to Auphonic's own conditions. Those allowances, prices and conditions may change, so users should consult the current information at:
This separation is important. Hartgen Consultancy and Desiree are not charging for Accessible Auphonic. The app is a free, accessible interface for using the Auphonic service. Whether you stay within Auphonic's free allowance or purchase processing time from Auphonic is entirely your decision.
Before a recording is uploaded, Accessible Auphonic asks for confirmation. The User Guide also explains what is sent online, how the API key is stored, how to remove it and where completed recordings are saved.
Who May Find Accessible Auphonic Useful?
Accessible Auphonic may be especially valuable to podcasters, musicians, trainers, teachers, broadcasters, oral historians, audiobook narrators, community groups and anyone who works regularly with spoken audio.
It may help the person who records a remote interview and struggles with one quiet participant. It may help someone preparing a set of lessons recorded in the same room. It may help a singer or podcast producer combining separate voice and music tracks. It may help a blind user who has heard terms such as loudness normalization or ducking but has never been given a patient explanation of what they mean.
It may also help experienced Auphonic users who already know the service well but want a faster desktop workflow with less web navigation.
Above all, it is intended for people who want to make thoughtful decisions about their audio without accessibility getting between them and the result.
Available Soon, and Completely Free.
Accessible Auphonic will be available from the Hartgen Consultancy website within the next few days in both installable and portable editions.
There will be no charge for the application.
This project began with one blind person's lifelong love of sound, her frustration with the tools available to her, and her decision to see whether she could build something better. It has grown through careful testing, collaboration, detailed documentation and real-world use.
We hope it will allow more blind people to approach audio processing with curiosity rather than apprehension. You do not need to begin by understanding every technical term. Begin with a recording. Identify the problem you can hear. Choose one change. Audition the result. Compare it with the original. Then decide what sounds better.
That is what Accessible Auphonic is really about: making sophisticated audio processing easier to reach, easier to understand and easier to control.
For JAWS users. For NVDA users. For anyone with a recording they want people to hear at its best.