AI podcast generator
Turn a topic, key points, tone, and speaker cast into an outline, editable transcript, generated performances, and a mixed podcast episode.
Open Podcast StudioPROGRESS
Podcast production
- Outline
- Transcript
- Audio
- Combine
Ready to generate…Not started
MULTI-SPEAKER PRODUCTION
Podcast Studio
Describe your episode, pick your speakers, and generate.
TOPIC & STYLE
Episode briefing
Source material & audience Optional
Script generation
Provider credentials and internal routing remain server-managed.
SPEAKERS · 1–4
Cast
1
Voice
2
Voice
3
Voice
4
Voice
Select 1-4 speakers below
YOUR EPISODE
Episode
- 1Describe your topic
- 2Pick speaker voices
- 3Generate & preview
♪
No episode yetGenerate your podcast to preview it here and in the player.
Export and send to Video Editor
PRODUCTION
Outline, transcript & clips
Not started
PRODUCTION
Outline, transcript & clips
Generate the outline to begin. Each stage is resumable.
Outline
Dialogue
0 linesHISTORY
Previous episodes
HISTORY
Previous episodes
LONG-FORM NARRATION
Audiobook Studio
Turn a manuscript into consistent, chaptered narration.
BOOK & MANUSCRIPT
Narration source
Narrator voice
YOUR AUDIOBOOK
Narrated book
- 1Add your manuscript
- 2Review chapters
- 3Narrate & master
♪
No audiobook yetYour generated narration will appear here and in the player.
STRUCTURE & PERFORMANCE
Chapters and narration
MASTER & EXPORT
Audiobook delivery
MASTER & EXPORT
Audiobook delivery
GETTING STARTED & SUPPORT
How To & FAQ
Learn the studio workflow and find answers about voices, local inference, audiobooks, podcasts, and export.
HOW TO USE THE STUDIO
From idea to finished audio
- 01
Choose a workflow
Start with a podcast topic, book manuscript, voiceover script, catalog voice, saved voice, or voice description.
- 02
Configure the voice
Select language, quality, delivery, and optional expert generation parameters.
- 03
Generate and review
Follow progress, review transcripts and takes, then adjust the inputs when needed.
- 04
Save or export
Reuse voices and personas, download the result, or prepare it for PixLab Video Editor.
ABOUT PIXLAB TTS STUDIO
Free local TTS, AI audiobooks, podcasts, and voiceovers
Start free with local TTS and an unrestricted in-browser text-to-speech workflow for supported models, or choose from 100+ local and hosted AI voice models. PixLab TTS Studio brings long-form audiobook narration, natural voiceovers, description-based voice design, and editable multi-speaker podcast production into one browser-based audio workspace. Create chapters, refine transcripts, reuse saved voices and personas, and prepare finished speech assets for PixLab Video Editor.
AI audiobook generator
Turn a manuscript into consistent narration with chapter detection, pronunciation guidance, audiobook metadata, and M4B, MP3, or WAV delivery.
Open Audiobook StudioAI voiceovers and voice design
Generate narration from a script or describe an entirely new voice by tone, accent, pacing, energy, age, and speaking character.
FREQUENTLY ASKED QUESTIONS
PixLab TTS Studio FAQ
What is PixLab TTS Studio?
PixLab TTS Studio is a free-to-start browser-based AI voice production workspace with 100+ models for local text to speech, chaptered audiobooks, voiceovers, description-based designed voices, and editable multi-speaker podcasts.
Is PixLab TTS Studio free to use?
PixLab TTS Studio is free to start, including supported local browser tools. Optional premium features, compute-intensive hosted models, and additional API usage may require credits or a paid plan.
What does unrestricted local TTS mean?
For compatible models, unrestricted local TTS means text-to-speech inference can run directly in the browser without sending the text to a hosted inference service or consuming hosted-model credits. Model, device, browser, safety, and lawful-use limits still apply.
Can text-to-speech inference run locally in my browser?
Yes. PixLab TTS Studio supports local in-browser inference for compatible TTS models and devices. Advanced or compute-intensive models can use secure hosted inference, so availability depends on the selected model and browser capabilities.
How many AI voices and text-to-speech models are available?
PixLab TTS Studio provides access to a growing catalog of more than 100 state-of-the-art text-to-speech models and voices covering different languages, accents, speaking styles, and production needs.
Can I generate a podcast with multiple AI speakers?
Yes. The Podcast Studio supports one to four speakers with individual roles and voices, plus outline generation, an editable transcript, audio generation, mixing, and episode history.
Can I create an audiobook from a manuscript?
Yes. Audiobook Studio accepts pasted manuscripts and prepared source files, organizes narration into chapters, preserves a consistent narrator and pronunciation guidance, and prepares chaptered M4B, MP3, or WAV output.
Can PixLab TTS Studio create voiceovers for videos?
Yes. The Voiceover workspace turns scripts into narrated audio with selectable voices, languages, delivery styles, output formats, and generation controls for video narration and other media.
How do I create a custom AI voice for my brand?
Open the Voice Design workspace and describe the voice in plain language — accent, age range, tone, pace, energy, and speaking style. The studio generates three comparable takes; preview each one, keep the best fit, and save it to your library for every future project.
Are designed AI voices original and safe for commercial content?
Yes. Every designed voice is a brand-new synthetic identity created from a text description and does not reproduce any real person. Designed voices stay private until you save them, and each voice in your library carries its own commercial-use eligibility flag.
Can I edit the generated podcast transcript?
Yes. Generated podcast dialogue is presented as an editable speaker transcript, so individual lines can be revised before selected audio is regenerated and combined.
Which text-to-speech controls are available?
PixLab TTS Studio provides Fast, Balanced, and Quality presets alongside controls for temperature, Top-K, Top-P, repetition penalty, maximum tokens, and additional intonation parameters.
Does PixLab TTS Studio work with PixLab Video Editor?
PixLab TTS Studio is designed as a companion to PixLab Video Editor, making it possible to prepare voiceovers, podcast audio, and timed speech assets for video projects.
Do I need to install software to use PixLab TTS Studio?
No desktop installation is required. PixLab TTS Studio is designed to run in a modern web browser with JavaScript enabled.
NARRATION & TEXT TO SPEECH
Voiceover
Turn a script into natural narration with block-level control over voices, delivery, and output.
VOICE SETTINGS
Voice and language
Voice
DELIVERY CONTROLS
Direction and emphasis
OUTPUT SETTINGS
Format & captions
Video Editor handoff
SCRIPT BLOCKS
Script
+
Add your first script block to begin.
- 1Add a block
- 2Pick its voice
- 3Generate takes
0 characters · ~0s · ~0 credits
♪
No audio generatedYour generated take will appear here and in the player.
TAKES
Block takes
Select a block to compare its takes.
CUSTOM AI VOICES
Voice Design
Create an original signature voice from a plain-language description — no recordings needed. Generate three takes and keep the one that fits your brand.
VOICE DESCRIPTION
Character and delivery
Voice recipe Tap attributes to build the description
Age
Accent
Tone
Pace
Style
Need inspiration?
Variants stay private until you save one. Saved voices appear in your Voice Library and every picker instantly.
TEXT INPUT
Preview script
♪
No variant generatedVoice previews will appear here.
Save result as Optional
PRESET & CUSTOM VOICES
Voice Library
Preview preset voices, manage saved custom profiles, and generate speech from either collection.
YOUR VOICES
Available voices
Atlas
Warm · Narration · Multilingual
DocumentaryCalm
Maya
Conversational · Podcast · English
NaturalFriendly
Theo
Authoritative · Documentary · English
DeepMeasured
Nora
Bright · Social video · Multilingual
EnergeticClear
Nova Designed
Designed voice · warm narrator
DesignedPrivate
Elena
Warm · Audiobook · Spanish
StorytellingSoft
GENERATE WITH SAVED VOICE
Quick generation
Selected voice
♪
No speech generatedThe selected voice output will appear here and in the audio player.
OPTIONAL PODCAST CHARACTERS
Character Profiles
Character Profiles are optional. They are only useful when a podcast needs a reusable identity, expertise, and speaking behavior attached to a voice; normal TTS and voiceovers do not need them.
VOICE SELECTION
Character voice
Select voice
CHARACTER DEFINITION
Identity and behavior
SAVED CHARACTERS
Character gallery
＋
Saved podcast character profiles will appear here.
AUDIO WORKSPACE & ACTIVITY
Projects & History
Continue production work, then search, replay, favorite, inspect, and manage generated audio.
Product launch conversation
4 speakers · 6 segments · Draft
Video Editor onboarding
12 blocks · 03:42 · Completed
Start a new project
Create a podcast, audiobook, voiceover, or designed voice.
SELECTED PROJECT
Revision history
Select a project to see its saved revisions.
GENERATION ACTIVITY
History
OutputVoice / typeDurationCreatedActions