The "I" narrator gets their own voice
In first-person promos, "I" is both narrator and character. AI tells narration from dialogue: narration uses a close storytelling voice, and spoken lines switch to an in-character delivery.
Novel promo videos live or die on the first three seconds and the pacing after that. One voice reading start to finish, and listeners swipe away. With this novel promo voiceover software you paste the copy, AI separates the "I" narrator from every character, gives each a voice and reads every line with feeling, and you export MP3 and line-timed SRT for CapCut.
Pay per character · Failed runs refunded · Alipay / WeChat Pay accepted
Sample D · The CEO who falls asleep to her voice
Chinese · Script excerpt · 2 characters

Heroine · narrator ("I")Tender
签下隐婚合约的第三个月,我才知道,他每晚都在听我录的睡前故事。
Three months into our secret-marriage contract, I found out he fell asleep to my bedtime stories every night.

CEO · Lu ChenTender
今晚的故事,换你当面讲给我听。
Tonight's story, tell it to me in person.

Heroine · narrator ("I")Whisper
陆总,合约里……没写这一条。
Mr. Lu, that's… not in the contract.

CEO · Lu ChenCalm
那就加上。
Then add it.
An excerpt from The CEO who falls asleep to her voice: the heroine tells the story in first person, and the CEO's lines switch to a different voice. That's the difference between a promo voiceover and a plain read-through.
CEO romanceShe signs a secret-marriage contract, then learns the CEO falls asleep to her voice every night.
40s
Sample D · The CEO who falls asleep to her voice
🇨🇳 Chinese40s
Script excerpt

签下隐婚合约的第三个月,我才知道,他每晚都在听我录的睡前故事。

今晚的故事,换你当面讲给我听。

陆总,合约里……没写这一条。

那就加上。
Cast of this episode
CEO's assistant · Zhou Min
Heroine · narrator ("I")
CEO · Lu Chen
Little brother · Xiao Zhou
Comic sidekick · Big Liu
Best friend · TangtangPromo accounts compete on posting speed and watch-through rate. The voiceover has to be fast, and it has to perform.
In first-person promos, "I" is both narrator and character. AI tells narration from dialogue: narration uses a close storytelling voice, and spoken lines switch to an in-character delivery.
Male lead, heroine, best friend, wicked mother-in-law: each gets a voice, so listeners know who's speaking without reading subtitles.
Pause before the twist, hit hard on the comeback, go soft when hearts flutter. Emotion tags apply line by line and make the hooks land instead of flattening them.
Paste a few thousand words and generate it in one go (up to 8,000 characters per run). For a series, keep a cast sheet and pick the same voice for each character every post, so daily posting never stalls.
Short-video promos, audio apps, authors promoting their own books and studios in bulk.
Hook them in three seconds: tag the opening lines Excited or Cold laugh, pause before the twist, hit hard on the comeback. The MP3 and SRT go into CapCut with your visuals.
Listeners there prefer a calmer read: switch the narrator to a radio or storyteller voice, soften the emotion tags, and the same copy becomes the audio version.
Paste the most gripping passage of a chapter, give the leads their own voices, and share the clip on your book page or in your community.
Several posts a day: each series keeps its cast list for the next post, billing is per character, and high volume suits a plan with more credits.
From copy to voiceover and subtitles, with no recording and no manual subtitle timing.
Paste the promo copy or a novel chapter. AI separates narration, dialogue and inner monologue automatically.
Listen to the narrator's voice and each character's, swap any that don't fit, and push the emotion of the opening lines to grab attention.
Export MP3 with line-synced SRT, add your visuals in CapCut and post. We don't render video; the vertical cut happens in your editor.
We deliver audio and subtitles, not video. For a vertical cut, drop the episode MP3 and SRT into CapCut and they line up in minutes.
One file per episode with the narrator and every character in script order, plus one per extra language, ready for the audio track.
One cue per line, starting and ending exactly where that line's audio does, UTF-8 encoded. In CapCut it becomes a ready subtitle track.
Picked several languages? Each one's MP3 and SRT download together in one ZIP, the easiest hand-off to an editor.
Create a 9:16 project and import the footage for the episode.
Drag the episode MP3 onto the audio track, starting at 0 seconds.
Import the matching SRT with the subtitle import; the cues fall into place on the timeline.
Preview line by line, adjust subtitle position and size, then export.
Picture and final export happen in your own editor: we don't render video or lip-sync. If the original footage has Chinese voices baked in, use a voice-free version or the music-and-effects track.
SRT example (format only, timings illustrative)
A promo voiceover isn't an audiobook and it isn't a drama. It has its own rhythm and tricks.
The problem
Most promos are first person: "I" narrates one moment and speaks lines the next. Read it in one voice and narration blurs into dialogue; switch voices naively and "my" lines get assigned to someone else.
How we handle it
AI decides whether each line is narration, dialogue or inner thought. The narrator's narration and dialogue share one voice with different deliveries, and every other character keeps their own voice.
The problem
The first three seconds decide whether listeners swipe away, and every few lines needs a hook. Read at an even pace with even emotion, and even great copy won't hold anyone.
How we handle it
Tag the opening hook as Excited or Cold laugh, let pauses open up naturally around twists, and the whole piece rises and falls with its emotion tags instead of running at one speed.
The problem
Promo accounts often post several pieces a day, each a few thousand words. Recording, editing and timing subtitles alone is more than one person can keep up with.
How we handle it
Long copy generates in one go, and each series saves its cast list, so the next piece reuses it without picking voices all over again.
The problem
Short-video platforms want fast pacing and big emotion, while audio-app listeners prefer a calmer read, so the same copy often needs two versions.
How we handle it
Switch the narrator to a different style (late-night radio, storyteller, soothing female voice), adjust the emotion tags across the piece and regenerate, and you have the other platform's version.
This is multi-voice software built for novel promos: paste promo copy or a chapter, AI separates first-person narration, dialogue and inner monologue, gives "I" and every character a voice and generates each line with emotion. You get MP3, line-timed SRT subtitles and a ZIP, billed per character, with failed lines refunded automatically.
Choosing novel promo voiceover software, what to look for, voices, deliverables and billing.
The key is whether it can split characters: in a good promo voiceover, the narrator and each character have different voices with real emotional range. Single-voice readers suit audiobooks better than promos. This novel promo voiceover software separates the "I" narration from dialogue and voices each.
Romance promos aimed at women often use a gentle, soothing female voice; male-oriented and suspense promos often use a magnetic radio-style male voice or a storyteller. Our library is sorted into narrators, male leads, female leads and villains, so preview before you pick.
No. Paste the copy and AI identifies the narrator and every speaking character. If a line is assigned to the wrong speaker, just change it by hand.
No. We deliver MP3, line-timed SRT subtitles and a multi-language ZIP, not rendered video. Create a 9:16 project in CapCut, import your visuals, put the MP3 on the audio track and import the SRT; the timings line up on their own.
Audiobooks aim for comfortable long listening at a steady pace; promos aim to hold attention, with a gripping opening and twists that hit. Line-by-line emotion tags and multi-character casting suit short, fast, hook-driven content.
Credits are charged on the characters you generate, narration and dialogue alike, and failed runs are refunded automatically. If you post a lot, choose a plan with more credits; see the pricing page.
Four things: can it split narrator and characters, can emotion be set line by line, can each character keep one fixed voice, and do subtitles come with per-line timing. The first two decide how it sounds; the last two decide how fast you can post daily.
Up to 8,000 characters per run, which usually covers a whole promo; split longer chapters. A new post in the same series keeps the characters already cast.
Yes. The same cast speaks 8 foreign languages, including English, Thai, Vietnamese and Indonesian, and each character keeps their voice, so a hit promo can travel to overseas platforms.
Failed lines regenerate in one click and their credits come back automatically. If a character's voice doesn't fit, swap it in the cast list and only that character's lines are re-voiced.
Still have a question? Open a ticket






Paste one promo script, analyse the cast for free, and hear the narrator and characters split apart. Pay per character; failed lines are refunded automatically.