Vokna is an AI Voice Studio that turns text into natural-sounding audio. Write your script in a rich text editor, assign speakers with different voices, and generate professional audio for podcasts, audiobooks, narrated content, and more.

The core idea: you write text, assign voices to speakers, and Vokna generates audio for each segment. You can then preview, edit, and download the result.

Understanding the hierarchy is key to using Vokna effectively:

Project > Section > Segment > Speaker + Voice

• Project: The top-level container. A project might be “Company Podcast” or “Audiobook Draft”.
• Section: A chapter or part within a project. Each section has its own editor and audio timeline.
• Segment: A block of dialogue assigned to a single speaker. Created in the editor using @mentions.
• Speaker: A voice persona with a name, color, and assigned voice. Speakers belong to projects.
• Voice: The actual AI voice model used for text-to-speech. From the library, or clone/design your own.

Every piece of audio needs a segment, and every segment needs a speaker with an assigned voice.

When you first open Vokna, a default project with one section and a “Narrator” speaker is created automatically.

1. 1Understand the auto-created structure
2. One project (rename it to something meaningful)
3. One section (“Untitled Section”)
4. One speaker (“Narrator”) with the default voice (Adam)
5. 2Write your content
6. Click in the editor and start typing
7. Use @SpeakerName to assign text to a speaker
8. Each @mention starts a new segment
9. 3Generate audio
10. Click “Voice All” to generate audio for all segments
11. Wait for processing (each segment shows a spinner)
12. Preview using the audio player at the bottom
13. 4Download
14. Use the download button to save as MP3 or WAV

Projects are created from the sidebar (folder icon). Each project contains sections and speakers.

Creating a project: Click the + button in the Projects panel, type a name, and press Enter.

Project limits:

• Free: 3 projects
• Starter: 15 projects
• Professional: Unlimited

Renaming: Click the project name in the sidebar or editor header to edit inline.

Deleting: Use the trash icon on the project row. This deletes all sections, segments, and audio permanently.

Sections are chapters or parts within a project. Each section has its own editor and audio timeline.

Every project must have at least one section. If a project somehow has zero sections, Vokna automatically creates an “Untitled Section”.

Creating a section: Expand a project in the sidebar, click the + button in the Sections tab, type a name, and press Enter.

Switching sections: Click a section name in the sidebar to load it in the editor.

Reordering: Drag sections using the grip handle to reorder them.

Segments are the core unit of content. Each segment is a block of text assigned to one speaker. Segments are created automatically in the editor.

How segments are created:

1. 1Type in the editor
2. 2Use @SpeakerName (or click the Speaker button) to assign a speaker
3. 3Type the dialogue text after the speaker badge
4. 4Start a new @mention to create the next segment

Minimum structure for audio generation:

• The segment must have non-empty text
• The segment must have a speaker assigned
• That speaker must have a voice assigned
• A draft revision must exist (auto-created)

Segment action dock: Click the speaker pill or double-click / highlight text within a segment to reveal the dock, which shows speaker info, revision history, voice instructions (paid tiers), and action options.

Revisions: Each segment can have multiple audio “takes”. Edit the text and create a new revision to try different versions without losing existing audio.

Speakers represent the voices in your project. Each speaker has a name, color, and an assigned voice from the library.

Default narrator: Every new project comes with a “Narrator” speaker using the Adam voice. You can rename, reassign, or delete this speaker.

Creating a speaker:

1. 1Go to the Speakers tab in the project sidebar
2. 2Click + to create
3. 3Enter a name and optional role
4. 4Pick a color (random or custom)
5. 5Select a voice from the library

Assigning speakers to segments: Type @ in the editor and select from the dropdown, or use the Speaker button in the toolbar.

Speaker limits: Free plan has 6 speakers per user. Paid plans have unlimited.

Unassigned segments: If a speaker is deleted, their segments become unassigned. You’ll need to reassign these to a real speaker before generating audio.

Vokna offers three types of AI voices:

Standard voices: Built-in voices available to all users. Browse them in the Library (audio lines icon in the sidebar). They use reference audio cloning for natural results.

Designed voices (Starter+): Create custom voices from text descriptions. Go to the Persona tab (sparkles icon), enter instructions like “A warm, friendly female voice with a British accent”, and Vokna generates a unique voice.

Cloned voices (Professional): Clone a voice from audio samples. Go to the Clone tab (fingerprint icon), upload reference audio, and create a voice that sounds like the source.

Voice preview: Click the play button on any voice card to hear a sample before assigning it.

Edit the speaker and select a different voice from the library. Existing audio will need to be regenerated.

1. 1Open the sidebar and go to the Speakers tab
2. 2Click the edit icon on the speaker
3. 3Click Select Voice and choose a new voice
4. 4Click Save
5. 5Choose whether to keep or clear existing audio recordings
6. 6Regenerate audio for segments that need the new voice

Audio generation is triggered by the “Voice All” button in the editor toolbar.

1. 1Write text with @speaker mentions
2. 2Click “Voice All” (shows count of unprocessed segments)
3. 3First click changes to “Confirm?” (safety check)
4. 4Second click starts generation
5. 5Segments show spinning indicators while processing
6. 6Completed segments show a checkmark

Credits: Audio generation uses credits (1 credit = 1 character). Free tier gets 10,000 credits/month.

Voice instructions (paid tiers): Add instructions like “speak slowly” or “add emphasis” to individual segments for finer control.

Failed segments: You are charged once when generation starts. If generation fails, retrying that segment is free and will not deduct additional credits. Click “Voice All” again to retry.

The audio player at the bottom of the editor provides full playback controls.

Controls:

• Play/Pause (center button with glow effect)
• Skip Forward/Back (navigate between segments)
• Repeat modes: Off, Continuous (play through all), Repeat (loop current)

Timeline: The section timeline shows a waveform view of all segments with word-level highlighting synced to playback, pause markers you can drag to insert silence, and segment boundaries as dividers.

Pause insertion: Drag pause markers on the timeline to add silence between words or sentences.

Use the download button (bottom right of the audio player) to export your audio.

Format options:

• MP3: Fast, smaller file, available on all tiers
• WAV: Higher quality, available on Starter+

Download options:

• Current clip: Download just the active segment
• All audio: Download all segments in the section concatenated as one file

Segment-level download: Click the download icon on individual segments in the timeline for single-clip exports.

1. 1Open the sidebar by clicking the folder icon in the top-left
2. 2Expand a project by clicking its name
3. 3Make sure the Sections tab is active (left tab)
4. 4Click the + button next to the Sections heading
5. 5Type a name for the section and press Enter
6. 6Click the section name to load it in the editor

1. 1Open the sidebar and expand a project
2. 2Click the Speakers tab (right tab)
3. 3Click the + button
4. 4Enter a name and optional role (e.g. “Interviewer”)
5. 5Pick a color (click the swatch to cycle or use the picker)
6. 6Click Select Voice and choose from the library
7. 7Click Create

1. 1In the Speakers tab, click the edit (pencil) icon on the speaker row
2. 2Click Select Voice
3. 3Browse or search the voice library, click a voice to select it
4. 4Click Save
5. 5If the speaker already has audio, you’ll be asked whether to clear recordings or keep them

1. 1Click in the editor where you want to start writing
2. 2Type @ and a speaker name will appear in a dropdown
3. 3Select a speaker from the list (or keep typing to search)
4. 4A colored speaker badge appears inline
5. 5Type your dialogue text after the badge
6. 6To start a new segment, type @ again and pick another speaker

1. 1Generate audio for a segment first
2. 2Click on the segment to open the audio player timeline
3. 3Drag a pause marker to the position where you want silence
4. 4The pause duration is stored and applied on download

1. 1Click the speaker pill or double-click / highlight text within a segment to open the action dock
2. 2Click the revision history icon (clock)
3. 3Click + to create a new revision
4. 4Edit the text in the new revision
5. 5Generate audio for the new revision
6. 6Switch between revisions using the revision list

The toolbar floats on the right side of the editor. It gives you quick access to the tools you use most: generating audio, adding speakers, opening the co-writer, and exporting your work.

The toolbar is always visible while you are editing a section. Each button opens its action directly or shows a small popover with more options.

The Voice All button (paper plane icon) generates audio for every segment that is ready. A badge on the button shows how many segments are waiting.

1. 1Click the paper plane icon
2. 2The button shows a checkmark and asks you to confirm
3. 3Click again to start generation

This two-step confirm prevents accidental generation. Your changes are saved automatically before generation begins.

Click the @ button on the toolbar to open the speaker menu. From there you can pick an existing speaker or create a new one. The same menu opens when you type @ in the editor.

Click the co-writer button (robot icon) on the toolbar to open the AI chat panel on the right. Click it again to close the panel. The button highlights when the panel is open.

The download button opens a popover with audio and text export options:

• Playback mode: Set repeat to off, repeat current clip, or play all clips
• Audio format: Choose MP3 (all plans) or WAV (Starter and above)
• Current clip: Download just the segment you are playing
• Download section: Export all audio merged into one file, or as a ZIP of individual clips
• Download text: Export your script as a .txt file, with or without speaker names

The script mode button (rows icon) switches the editor between two views:

• Inline view: Segments flow as a continuous block of text
• Script view: Each segment is shown as a separate card

Pick the view that fits how you work. The setting is saved per section.

The clear button (eraser icon) removes all content from the current section. Click it once to open a confirmation, then click Yes to confirm.

This deletes every segment and its audio from the section. You cannot undo this, so make sure you want to start fresh.

The co-writer is an AI assistant accessible from the right sidebar. It helps you brainstorm, outline, and draft content.

Features:

• Suggestion chips: “Help me outline this”, “Write a first draft”, “Create new characters”
• Context-aware: It knows your current section content and speakers
• Can create and edit segments directly in the editor
• Streaming responses with reasoning display

1. 1Click the co-writer toggle in the toolbar or sidebar
2. 2Type a prompt or use a suggestion chip
3. 3Review AI-generated segments (green preview cards)
4. 4Accept or reject changes

Each project can have up to 50 co-writer conversations. Switch between them in the chat panel.

Customize the editor from the Settings panel (gear icon in the sidebar).

Theme: Choose from 15+ themes (light and dark modes). Live preview shows how your content will look.

Typography: Font size (14-24px), font family (Inter, Sans, Serif, Mono, Menlo, Roboto, Open Sans), line height and paragraph spacing.

Speaker mentions: Show/hide speaker colors, badge shape (none, sm, md, lg, full rounded).

Behavior: Auto-save interval (off, 30s, 1m, 5m), spell check language, word/character count display.

All settings sync to your account and persist across sessions.

Vokna offers four tiers:

Free ($0/month): 10,000 credits/month, 3 projects, 6 speakers, MP3 export, standard voices.

Starter ($5.99/month): 50,000 credits/month, 15 projects, unlimited speakers, MP3 + WAV export, designed voices, voice instructions.

Professional ($19.99/month): 250,000 credits/month, unlimited projects and speakers, voice cloning, priority support.

Studio (Custom): Unlimited credits, all Professional features, HIPAA/SOC2 compliance.

Credits: 1 credit = 1 character of text. Credits reset monthly. Paid tiers support overage billing.

Click your profile > Billing to access the customer portal. From there you can upgrade, downgrade, or cancel your subscription. Your plan stays active through the end of the current billing period.

• Use sections to organize long content: Break scripts into logical sections (chapters, scenes, episodes) rather than one massive block.
• Name your speakers clearly: “Narrator”, “Interviewer”, “Guest” are clearer than “Speaker 1”.
• Preview voices before assigning: Click the play button on voice cards to find the right fit.
• Use voice instructions for nuance: On paid tiers, add instructions like “pause after each sentence” or “speak with excitement” for better results.
• Create revisions instead of overwriting: When experimenting with different phrasings, create new revisions rather than editing existing text. This preserves your best takes.
• The default Narrator is a starting point: Feel free to rename, reassign, or delete the auto-created Narrator speaker.
• Use the co-writer for first drafts: Let the AI generate an outline or draft, then refine it yourself.
• Check the timeline for pacing: The waveform view helps you spot segments that are too long or need pause insertion.

If the waveform shows a clip but playback fails:

1. 1Verify the browser tab isn’t muted (check the browser tab’s speaker icon).
2. 2Check system audio output device and volume (Settings > Sound).
3. 3Refresh the editor page and try playback again.
4. 4Try a different supported browser (Chrome or Edge recommended).

If an error message appears, note the exact text and contact support with the project name, version number, and browser details.

If changes to the timeline don’t persist:

1. 1Confirm you saved the project after making the edit (save indicator in toolbar).
2. 2Check the project version isn’t locked or in a read-only state.
3. 3Reload the editor to ensure the latest state is loaded.

If edits still don’t persist, capture a screenshot of the timeline and any error text, then contact support with the project name and version.

Changing a speaker’s voice does not automatically regenerate existing audio. You must regenerate the segment(s) to apply the new voice.

1. 1Open the Speakers tab in the sidebar and confirm the new voice is assigned.
2. 2In the editor, select the segment(s) that need the new voice.
3. 3Click “Voice All” (or regenerate the individual segment) to create audio with the updated voice.

If you want to keep the old takes, create a revision first (clock icon in the segment toolbar), then generate the new revision with the updated voice.

Common causes and fixes:

• Wrong voice: Verify the speaker has the correct voice assigned and that you regenerated after any change.
• Missing voice instructions: Voice instructions (paid tiers) only apply to the segment they’re added to. Check the segment toolbar > voice instructions.
• Pronunciation issues: Try spelling out numbers, acronyms, or unusual words phonetically. For persistent issues, note the exact script excerpt and direction tags used.
• Stale revision: Make sure you’re playing the latest revision. Switch revisions via the segment toolbar clock icon.

To isolate the cause, generate a short test segment with the same settings. If the issue persists, capture the script excerpt, direction tags, and any error/warning text, then contact support.

If export fails:

1. 1Note the export format selected (MP3 or WAV) and any error message shown.
2. 2Try exporting a single segment first (segment download icon in timeline).
3. 3If single-segment export works, try “All audio” again.

WAV export requires Starter tier or above. If the error persists, record the exact error message, project name, version, and browser, then contact support.

When a segment fails to generate, a Retry button appears. Clicking Retry does not consume additional characters — the original generation attempt was already charged. Retrying reuses the same character allocation.

Retry creates a new revision (a new take) rather than overwriting the failed attempt, so you can compare versions. If Retry fails again, check character balance (Settings > Billing) and try a shorter segment to isolate the issue.

Yes. You are charged 1 credit per character when you click Voice All. If your monthly character balance reaches zero, new generation (including Retry) will be blocked until the balance resets or you upgrade.

Check your remaining credits in the sidebar (credit counter) or Settings > Billing. Free tier resets monthly; paid tiers support overage billing.

Vokna is optimized for Chrome and Edge (Chromium-based).

• Safari: Web Audio API differences may cause playback/timeline issues.
• Firefox: Some editor interactions (drag-drop, audio context) may behave inconsistently.

If you encounter issues in other browsers, switch to Chrome or Edge as a first troubleshooting step.