User Guide for HeardThat Plus

Table of contents

1. Introduction to HeardThat Plus

Welcome to HeardThat Plus, a web application that lets you improve audio quality, speech intelligibility, and create accurate transcripts.

What HeardThat Plus does

HeardThat Plus offers two primary functions:

1. Noise removal: Our AI-powered system removes background noise from audio recordings.
2. Transcription: HeardThat Plus generates written transcripts from your audio files.

In the following we will use the term "project" to describe the uploaded media file, custom vocabulary, and all the outputs from HeardThat Plus.

Key features

• Noise removal: Cleans up audio recordings, even those with significant background noise.
• Transcription: Converts spoken words in your audio files into written text, with speaker labels for who said what, when. Our noise removal technology improves the accuracy of transcripts from noisy recordings.
• Custom vocabulary: You can add specialized terms or names to improve transcription accuracy for your specific content.
• Transcript editing: Our interface allows you to review and edit transcripts, with several tools to make this process efficient.

Who can benefit

HeardThat Plus can be useful for a variety of users, including:

• Content creators looking to improve audio quality or add captions
• Journalists working with interview recordings
• Researchers transcribing focus groups or field recordings
• Anyone needing to clean up audio or create transcripts from noisy recordings

This guide will walk you through how to use HeardThat Plus, from uploading your audio files to editing your final transcripts to downloading subtitles. If you have any questions along the way, please try our experimental chatbot, or contact our support team at support@singularhearing.com.

2. Getting started

HeardThat Plus is a web-based application that's easy to access and use. Here's how to get started:

Accessing HeardThat Plus

1. Open your web browser. HeardThat Plus works on all common browsers, both on desktop and mobile devices.
2. Navigate to the HeardThat Plus website.

Creating an account

1. On the login page, you have two options for creating an account:
   • Sign up with your email address and create a password
   • Sign in with your Google or Apple account
2. New users receive 100 free credits to try out the service.

Logging in

Once you've created your account, click Login in the navigation bar to get to the login page.

Your first use

1. After logging in, you'll be taken directly to the Upload page.
2. Here, you can immediately start using HeardThat Plus by uploading an audio file.
3. Choose which tasks you want to perform on your audio:
   • Clean (remove noise)
   • Transcribe
   • Both clean and transcribe
4. Click the button to start processing.

Note that when you hover over a task button, the estimated amount of credits to execute the task will be shown. Your total remaining credits are displayed next to your profile picture in the upper right.

Enabling notifications

The first time you use HeardThat Plus, you might be asked (depending on your browser settings) if you want to enable notifications. We recommend allowing notifications because:

• Processing audio can sometimes take a few minutes, especially for longer files.
• With notifications enabled, you can work on other tasks and be alerted when your results are ready.

That's it! You're now ready to start using HeardThat Plus to clean your audio and create transcripts. The following sections will guide you through each feature in more detail.

3. Dashboard

After logging in, you'll be directed to your Dashboard. This is your central hub for managing projects. Here's how to navigate and use the Dashboard effectively:

1. Available storage: At the top of the Dashboard, you'll see a progress bar indicating your available storage. This shows how much of your allocated storage you've used and how much is still available. If you are running out of space, delete old projects that you don't need any more.
2. Action buttons:
   • Action: This dropdown allows you to perform bulk actions on selected files, such as downloading or deleting multiple files at once.
   • Sort by: Use this to organize your files by upload time or file name.
3. Project list: The main part of the Dashboard displays a table of your files with the following columns:
   • Name: Shows the filename of each audio file.
   • Length: Duration of the file in HH:MM:SS format.
   • Date: Displays when the file was uploaded.
   • Status: Indicates the current state of the file using icons:
     • speaker icon: Cleaned audio is ready
     • document icon: Transcript is ready
     • rotating cog: Task is in process
     • X: Task failed
   • Actions: For each project, you can:
     • Download: Access various versions of your audio file (original, cleaned, mixed) or transcript
     • Preview: Open the file for detailed reviewing and transcript editing
     • Delete: Remove the project from your account
4. Pagination: If you have many projects, use the pagination controls at the bottom of the table to navigate through the list.
5. Selecting projects: You can select multiple projects by checking the boxes next to them. This allows you to perform bulk actions (see Action Buttons above).

Your Dashboard is dynamically updated so as you process new files or make changes, it will reflect these updates in real time, ensuring you always have the most current view of your projects.

4. Uploading and processing audio

Supported file formats and sizes

HeardThat Plus supports the following audio and video formats (depending on your platform and browser):

• Audio: MP3, WAV, FLAC, M4A
• Video: MP4, MOV, AVI, MKV, FLV, WMV, WEBM

The maximum file size is 500 MB.

For video files, the audio is extracted automatically and processed. After processing, you have the option to download the video file with the original audio replaced with the cleaned audio, or whatever mix you have selected.

Uploading files

1. Navigate to the Upload page.
2. You can upload files in two ways:
   • Drag and drop your file(s) into the designated area.
   • Click on the upload area to open a file browser and select your file(s).
3. You can upload multiple files at once, up to a maximum of 20 files.
4. Once you've uploaded your file(s), their names will be displayed in the upload area.

Choosing processing options

After uploading your file(s), you have three processing options:

1. Clean: Removes noise from your audio.
2. Transcribe: Creates a transcript of your audio, including labels for the speakers.
3. Clean & transcribe: Performs both cleaning and transcription.

Click the corresponding button to start the processing. You will be taken to the Dashboard where you can monitor progress.

Note that if you uploaded multiple files the same processing will be applied to each.

Understanding credit usage

Credit usage is based on the duration of your audio and the selected processing option:

• Clean: 10 credits per hour
• Transcribe: 20 credits per hour
• Clean & transcribe: 30 credits per hour

The credit cost will be displayed when you hover over each option button. For files shorter than a certain duration, a minimum of 1 credit will be charged.

Custom vocabulary

Before processing, you can add custom vocabulary to improve transcription accuracy:

1. In the "Add your custom vocabulary here" input field, enter specialized terms or names.
2. Separate multiple terms with commas. You can also specify short phrases. Example: "Emilie, HeardThat Plus, SixEaster" (without the quotation marks).
3. This vocabulary will be used to output the preferred spelling in the transcript.

Processing

1. After selecting your processing option, the system will begin processing your file(s).
2. You can see the status of the processing in the Dashboard.
3. Once complete, you'll receive a notification (if enabled) and an email. The notification icon at the top of the Dashboard will get a red dot. If you click the notification icon you can click the processed item to be taken directly to the Preview page to listen to it.

5. Audio cleaning features

HeardThat Plus offers powerful audio cleaning capabilities to remove noise from your recordings. Here's how you can make the most of these features:

How audio cleaning works

HeardThat Plus uses advanced AI algorithms to separate speech from background noise in your audio files. The system discards the noise to create a "cleaned" version of your audio. This process not only improves speech intelligibility but also provides greater accuracy of transcriptions.

Previewing cleaned audio

After your file has been processed, you can listen to the original and cleaned versions of your audio:

1. From your Dashboard, click the "Preview" button for the project you want to review.
2. In the Preview page, you'll see two waveforms:
   • The top waveform represents your original audio.
   • The bottom waveform shows the cleaned audio.
3. Use the play/pause buttons near the Original and Cleaned tracks to hear just that track.
4. Use the play/pause button at the left to listen to a mix of the two audio tracks. The mixture can be adjusted with the slider at the left. You might want to mix in a bit of the original to make it sound more natural or to include some ambient sound.
5. The waveforms are synchronized, allowing you to visually compare the original and cleaned versions.
6. You can click anywhere on either waveform to jump to a specific part of the audio.

Adjusting mix ratio

HeardThat Plus provides a unique feature that allows you to adjust the mix between the original and cleaned audio:

1. On the Preview page, you'll find a vertical slider labeled "Mixing ratio" (or horizontal on smaller screens).
2. This slider controls the balance between the original and cleaned audio:
   • Sliding towards the top (or left on mobile) increases the volume of the original audio.
   • Sliding towards the bottom (or right on mobile) increases the volume of the cleaned audio.
3. As you adjust the slider, the audio playback updates in real-time, allowing you to find the optimal balance.
4. The mixing ratio is displayed as a percentage, indicating the proportion of original to cleaned audio.
5. You can use the play/pause buttons next to each waveform to listen to listen to just that track.

This mixing feature is particularly useful when you want to preserve some of the original audio ambiance while still benefiting from noise reduction. It allows you to find the perfect balance between noise reduction and maintaining the natural feel of your recording.

6. Transcription features

How transcription works

HeardThat Plus offers advanced transcription capabilities to convert spoken words into text. The system analyzes your audio file, identifying words, punctuation, and speaker changes. It works in tandem with audio cleaning to improve the accuracy of even noisy recordings.

Using custom vocabulary

There are many times when you know some words that might confuse any automatic speech recognition system. For example, less common spellings of names ("Jon" versus "John"), product names ("HeardThat Plus"), regional dialect ("yinz"), industry jargon, and so on. You can give HeardThat Plus a list of such spellings to use for your audio transcription.

1. At the time of uploading, enter custom terms or names in the "Add your custom vocabulary" field.
2. Separate multiple entries with commas (e.g., "Emilie, HeardThat Plus, SixEaster").

The system will prioritize these spellings during transcription.

Viewing transcripts

Once your audio is processed, you can view and edit the transcript:

1. From your Dashboard, click “Preview” on the desired project.
2. The transcript appears below the audio waveforms.
3. Each word is timestamped and linked to the corresponding audio.
4. Click on any word to jump to that point in the audio.
5. Alternatively, click on the audio waveform to move to that point in the transcript.
6. Use the search function to find specific words or phrases in the transcript.
   • Searches are case-insensitive.
   • You can use "*" as a wildcard match. Examples of matches:
     • plan* → planning
     • *anning → planning
     • *ann* → planning
     • *an*n* → Janine?, planning
   • There are buttons and shortcut keys for "Find previous" (Shift-n) and "Find next" (n).
7. NOTE: You can use the spacebar to play/pause the audio.

Editing transcripts

Editing transcripts is a little different from editing an ordinary text file because:

1. There are labels with speaker names.
2. Each word has an associated time.

There are just a couple of concepts you need to know. We explain those next. Then we will introduce several convenience features that make editing fast and easy.

Editing transcripts: the basics

Edit words:

• Select the word using the mouse or arrow keys.
• Enter edit mode by either pressing Enter or double clicking.
• To finish editing, press Enter or Esc, or click outside the edit area.

Change a speaker name:

• Right-click on the speaker label.
• Choose “Edit speaker name”.

Insert a speaker label before a word:

• Right-click on the word.
• Choose “Add speaker name”.

Move a speaker label:

• Click on it and drag-and-drop to the desired location.

Remove a speaker label:

• Right-click on the label.
• Choose “Delete speaker name”.

Undo/redo:

• Use the circular arrow buttons to the right of the Search button.

Editing transcripts: convenience features

Playback speed:

• Click the “Review” button to toggle it on. (It is in the toolbar above the transcript.)
• Choose the playback speed from the dropdown (0.25x to 2.0x).

Change case:

• Select a word or speaker label.
• Press the “c” key repeatedly to cycle through upper, lower, and sentence case.

Fix punctuation:

• Select a word.
• Type a punctuation key like comma, period, question mark, etc.
• If there is no punctuation at the end of the word, it will be added.
• If there is already punctuation, it will be replaced.
• If the punctuation is there and is the same as the key, it will be removed.

Quick speaker name operations:

• When you create a new speaker name it is automatically assigned a number.
• You can see the number in the dialog box when you add or edit a speaker name.
• Change name: Select a speaker label and type a number to change to the associated name.
• Insert label: Select a word and type a number to insert a label with the associated name.
• NOTE: The number shortcuts are only available for names that have been added.

Merge words:

• For example, change “Heard That” to “HeardThat”.
• Select the first word.
• Shift-click on a word to the right or left.
• All the words in between will be merged into a single word.

Speaker labeling

HeardThat Plus automatically identifies different speakers in your audio:

1. Speakers are labeled in the transcript (e.g., "SPEAKER_00", "SPEAKER_02").
2. To edit speaker labels:
   • Right-click on a speaker label.
   • Choose "Edit speaker name" from the context menu.
   • If you want to change all occurrences of that speaker label, choose "Select all matching labels".
   • Enter the new name and click "Save".
3. You can also insert speaker labels:
   • Right-click on a word where you want to insert a label.
   • Choose "Add speaker name" from the context menu.
   • Enter the new speaker's name and click "Save".
4. To remove a speaker break:
   • Right-click on the speaker label you want to remove.
   • Choose "Delete speaker name" from the context menu.
5. You can drag and drop speaker labels to fix errors.

Additional transcript editing features:

• Undo/Redo buttons allow you to revert or restore recent changes.
• An auto-scroll option keeps the transcript in sync with audio playback.
• Review mode is available to let you play back at different speeds. In Review mode, you can edit the transcript but you can't change the audio mix.

7. Working with results

Once your project has been processed, HeardThat Plus offers various ways to use and share your results.

Downloading audio and video

1. From the project Preview page:
   • Click the download button (arrow icon) next to the "Original" or "Result" waveform.
   • For the original audio, you'll download the unprocessed file.
   • For the result, you can choose between:
     • Cleaned audio (noise removed)
     • Mixed audio (based on your custom mix ratio)
   • Note: If the leveled option has been selected ("Level" button), then resulting audio will be leveled. This makes speaker segments more evenly loud across the recording. So for example, for an interview or meeting where one voice is loud and another is quiet, Level will even things out.
2. From the Dashboard:
   • Click the download button in the "Actions" column.
   • Select the version you want to download from the dropdown menu.

For video projects, you have additional options:

• Download the video with the cleaned/mixed/leveled audio replacing the original audio

Exporting transcripts

HeardThat Plus allows you to export your transcript in various formats:

1. From the project Preview page:
   • Locate the download button (arrow icon) in the transcript section.
   • Click to open a dropdown menu with different export options.
2. Choose from the following formats:
   • TXT: Plain text format
   • SRT: SubRip Subtitle format, commonly used for video subtitles
   • VTT: WebVTT format, used for web video subtitles
   • DOCX: Microsoft Word document
   • PDF: Portable Document Format

Sharing results with team members

HeardThat Plus makes it easy to collaborate and share your projects:

1. From the project Preview page:
   • Click the "Share" button.
   • A popup will appear with icons for viewing and editing.
   • When you click the icon, it will copy the share link to the clipboard.
2. You can send the copied link to team members. Anyone with the link will have access, so be careful how you share.
3. For recipients to access the project through this link, they will need a HeardThat Plus account. (It's free.)
4. Shared projects allow recipients to:
   • Listen to the original and cleaned audio
   • Change the mixing setting
   • Toggle the "Level" option
   • View and potentially edit the transcript (depending on permissions)
   • Download audio, video, and transcript files

The sharing link can be deleted at any time from the Dashboard or a Preview page. If deleted, the link will no longer work for anyone you had shared it with. If you want to restore sharing, you can create and share new links in the Dashboard or Preview page.

8. Advanced features

Video processing

HeardThat Plus is capable of processing video files in addition to audio files. Here's what you need to know:

1. Supported formats: MP4, MOV, AVI, MKV, FLV, WMV, WEBM (depending on platform and browser).
2. When you upload a video file, the system automatically extracts the audio for processing.
3. After processing, you have two options:
   • Download the cleaned audio separately.
   • Download the video with the original audio replaced by the cleaned audio.

To download the processed video:

1. Go to the Preview page for your video project.
2. Click the download button in the "Actions" column.
3. Choose "Video with cleaned audio" (or "mixed audio") from the dropdown menu.

This feature is particularly useful for content creators who want to improve the audio quality of their videos without having to open it up in their video editor and manually replace the audio.

Leveling audio

Audio leveling is an advanced feature that helps to balance the volume of your audio, making it more consistent throughout the recording and achieving an overall level that will be consistent across different recordings. This is especially useful for recordings with varying volumes, such as interviews or podcasts with multiple speakers.

To use the leveling feature:

1. On the Preview page, look for the "Level" button next to the "Cleaned" waveform.
2. Click the "Level" button to toggle leveling on or off.
3. When leveling is on, the button will be highlighted, and you'll see a new waveform representing the leveled audio.
4. You can now listen to and compare the original, cleaned, and leveled versions of your audio.

The leveling feature works in conjunction with the cleaning process, providing you with audio that is both noise-free and volume-balanced.

Advanced audio mixing

HeardThat Plus offers an advanced mixing feature that allows you to blend the original and cleaned audio to your preference. This is useful when you want to preserve some of the original ambiance while still benefiting from noise reduction.

Here's how to use the advanced mixing feature:

1. On the Preview page, locate the vertical slider labeled "Mixing ratio" (it may be horizontal on smaller screens).
2. Adjust the slider to change the balance between original and cleaned audio:
   • Moving the slider up (or left on mobile) increases the proportion of original audio.
   • Moving it down (or right on mobile) increases the proportion of cleaned audio.
3. The current mix ratio is displayed as a percentage.
4. As you adjust the slider, the audio playback updates in real-time, allowing you to hear the changes immediately.
5. Use the play/pause buttons next to each waveform to listen to only the original or only the cleaned audio.
6. Once you've found your preferred mix, you can download this custom mixed version of your audio.

Remember, you can always download both the fully cleaned and original versions separately if needed.

9. Account management

Managing your HeardThat Plus account is straightforward and can be done entirely through the user interface.

Updating profile information

To update your profile information:

1. Click on your profile picture in the top right corner of the page.
2. Select "My profile" from the dropdown menu.
3. On the Profile page, you can update the following information:
   • Name
   • Email address
   • Password
   • Subscription level
4. After making changes, click the "Save Changes" button at the bottom of the form.

Note: If you change your email address, you'll need to verify the new address before the change takes effect.

Changing password

For security reasons, it's a good idea to change your password periodically. Here's how:

1. Go to the Profile page as described above.
2. Click on the "Change password" tab or button.
3. Enter your new password twice to confirm.
4. Click "Update Password" to save the changes.

If you've forgotten your password:

1. Click "Forgot Password" on the login page.
2. Enter the email associated with your account.
3. Follow the instructions in the password reset email you'll receive.

Managing subscription and credits

HeardThat Plus operates on a credit-based system. Here's how to manage your subscription and credits:

1. From your profile dropdown, select "Subscription & Credits" or "Billing".
2. On this page, you'll see:
   • Your current subscription plan
   • The number of credits you have available

To purchase more credits:

1. Click the "Buy credits" button.
2. Select the credit package you want to purchase.
3. Follow the prompts to complete your payment.

To change your subscription:

1. Click "Change Plan" or "Upgrade Subscription".
2. Select the new plan you want.
3. Confirm the change and, if necessary, provide payment information.

To view your billing history:

1. Look for a "Billing History" or "Invoices" section on the Subscription page.
2. Here you'll find a list of all your past transactions.

Remember, you can always see your current credit balance in the top right corner of the Dashboard. If you're running low on credits, you'll receive a notification prompting you to purchase more.

If you have any issues with billing or credits, please contact our support team at support@singularhearing.com.

10. Pricing and subscriptions

For the most up-to-date information on our pricing and subscription options, please visit our Pricing page.

11. Additional sources of information

Contacting support

You can activate our live chat support by clicking the icon in the lower right corner.

Alternatively, send email to support@singularhearing.com.