Free
Script intelligence that reads your format
Paste your script. The parser detects speakers, SFX cues, music tags, and scene markers — and confidence-scores every line so you only review the ones it was unsure about.
On air — pixeLantern apps
Scripts become
Scripts become
podcasts.
flexVox turns a written script into a fully produced, multi-voice podcast. Paste dialogue, assign AI voices to every character, generate speech, sound effects, and music — then mix and export, all on your iPhone or iPad.
- iOS 26.4+
- iPhone & iPad
- Native SwiftUI
- Demo mode included
Script Production Export
00:00.0 · idle
ALEX: You hear that? The whole city is on a delay.
JORDAN: [whispers] Three seconds behind the lightning.
[SFX: distant thunder (3s) @underlay volume=0.4]
ALEX: [laughs] Promise me we'll never count down together.
[Music: slow tape-piano hush (12s) @underlay loop]
- ABriar — multilingual v2
- JRoan — turbo v2.5
A multi-speaker script lands. The parser detects ALEX, JORDAN, SFX, and music cues automatically.
Auto-Cast picks a voice for every character. Fine-tune stability, style, and speed.
Dialogue, SFX, and music are mixed with auto-ducking. Export to your podcast host.
The thesis
“Most TTS tools start with a single text box.
“Most TTS tools start with a single text box.
flexVox starts with a script.”
Every other text-to-speech app asks you to paste one paragraph for one voice. That's fine for a notification. It's not fine for a podcast, a table read, or an audio drama. flexVox preserves speaker, SFX, and music information through the entire pipeline — so what you get back actually sounds like a conversation.
No web views. No Electron. No accounts to start. Bring your own ElevenLabs key when you're ready for real audio — or stay in demo mode and explore every screen for free.
HOST: Welcome back to the show.
HOST: [warmly] Today, we're talking about timing.
[SFX: rain on windows (5s) @underlay volume=0.3 loop]
GUEST: [laughs] You picked the right week.
[Music: warm rhodes intro (10s)]
From a paste to a finished M4A.
Three tabs — Script, Production, Export — and one continuous workflow. Jump back and forth freely. flexVox remembers where you were.
- 01
Paste
Drop in a multi-speaker script. The parser detects HOST:, [Host], standalone names, and your SFX and music tags across common formats.
Script tab · auto-parse + confidence scoring
- 02
Cast
Auto-Cast picks a voice for every speaker. Or open the library and audition voices yourself — the session auto-advances to the next speaker.
Voice Mapping · Auto-Cast All + AI archetype banner
- 03
Generate
flexVox produces dialogue, sound effects, and music in one pass. Failed turns are retried automatically. Cancel anytime — already-generated audio is preserved.
Generation Status · resumable + resilient
- 04
Polish
Regenerate any line as a new variant. Adjust pauses. Mark music as underlay. LUFS-aware auto-ducking sits the bed at the right level on its own.
Post-Production · variants, ducking, flags
- 05
Export
Pick an export preset — Apple Podcasts, Spotify, YouTube, Broadcast, or Custom LUFS. Get an M4A, plus SRT, VTT, JSON, or plain-text transcript.
Export sheet · Cmd+E
Six headline features.
The whole production line. The free tier covers the full script-to-audio path; Studio adds the professional edge.
Free
A voice for every character
Browse the ElevenLabs catalog inline, audition voices instantly, and tune stability, style, and speed per speaker. Auto-Cast picks distinct voices for your whole cast in one tap.
Studio
SFX & music written into the script
Type [SFX: thunder (3s) @underlay] and hear it happen. flexVox generates, ducks, and places ambient beds and underlay music without leaving the editor.
Free
Regenerate one line. Keep everything else.
Each take is saved as a variant. Compare side-by-side, pick the best, and discard the rest. Your other audio never gets touched.
Studio
AI script writing that knows v3 tags
Describe an episode and get a production-ready draft. The engine knows 62 ElevenLabs v3 tags — [whispers], [laughs], [trying to sound brave] — and places them where they fit.
Free
Follow along, word by word
The Export view is a live teleprompter. Words highlight in time with the audio. Tap any line to jump there. Spot a mispronunciation without scrubbing.
- 62 v3 expression tags
across 4 categories - 2,000ch per dialogue batch
10 voices · auto-split - −16LUFS target loudness
Apple Podcasts preset - 0.5–30s SFX duration
Music up to 600s - 5 project templates
Interview, drama, news, narration, true crime
If you write dialogue, this is your studio.
flexVox is the tool we wished existed when we wanted to hear a script before we recorded it.
Run a narrative show but can't afford a cast? Write the parts. flexVox performs them — and lets you iterate without rebooking talent.
Generate a table read of your script with distinct character voices. Swap a voice, re-read a scene, share a rough audio draft in a single afternoon.
Build dialogue-based scenarios and role-plays. When your script updates, regenerate the lines that changed — keep everything else.
flexVox is not a general-purpose audio editor.
It does one thing well: turn a multi-speaker script into produced audio. Here's where it fits.
flexVox Generic TTS Desktop DAW Web AI tools
Multi-speaker per project Each character → separate voice Usually single-voice No built-in TTS Varies
Script parsing Auto speaker detection · confidence Manual entry N/A Varies
SFX & music generation Inline tags generate audio Not available Manual import Sometimes separate
Per-line regeneration Variants + side-by-side compare Not available Manual editing Limited
Platform Native iOS Mixed Desktop / iPad Browser
Try without an account Demo mode · zero commitment Usually requires login Fully offline Requires internet
flexVox depends on ElevenLabs for voice generation. Audio quality and available voices come from that service. We add the script intelligence, the production workflow, and the native mobile experience.
On air
Pull up a script. We'll cast it.
Free to download. Free demo mode. Bring your own ElevenLabs key — or just explore the workflow first.