01 / INPUT
Upload the audio
Your recorded voiceover, podcast episode, lecture, or audiobook chapter. Batch upload works too.
episode-042.wav · 01:40:12
Forced alignment · not transcription
You already have the script.We only find the timing.
Upload your voiceover and the exact words you want on screen. We align them to the audio — word by word — and return a clean SRT. No re-transcription. No misheard names. No rewriting.
- No credit card
- Clean SRT on the free run
- Files up to 4 hours
episode-042.wav·script-v7.txt
Ready
01INPUTDrop your audio and script to begin
AUDIODrop audio fileWAV · MP3 · M4A · MP4 · ≤ 4 h
SCRIPTDrop script fileTXT · DOCX · MD · keeps line breaks
02ALIGNMENTWaiting for input
Your script — locked to audio
So the first thing I tell every client isdon’t touch it for six weeks.I know that sounds impossible,but that’s exactly why most peoplenever find out what their hairis actually capable of doing.
03OUTPUTSRT · VTT · TXT
SRT · output0 / 6
episode-042.srt6 cues · 1.2 KBDownload
0:00 / 0:26
- Throughput
- 100minAligned in about a minute — no timeline dragging
- Rate
- $7/100minCredits never expire. No subscription needed
- Max file
- 4hrMaximum file length, with no surcharge
- Retention
- 24hrUntil your files are permanently deleted
How it works
Three steps. No timeline dragging.
This is a forced-alignment tool, not a transcription tool. The difference is why your output is exact.
02 / SCRIPT
Paste the exact script
Whatever you want burned on screen. Line breaks are respected; your terminology is never altered.
“So the first thing I tell every client is — don’t touch it.”
03 / OUTPUT
Download the SRT
Every line timed to the millisecond it’s spoken. Import straight into your editor or YouTube Studio.
.SRT · .VTT · .TXT
Under the hood
What happens inside the aligner
Four stages, no language model in the middle. Your words are never re-derived from the audio — they are located in it.
Stage 01
Audio in
16 kHz mono, cut into 20 ms frames. Voice activity detection skips silence.
Stage 02
Phoneme posteriors
Each frame is scored against the phonetic alphabet — not against English words.
Stage 03
Script scorer
Your script is converted into the same phonetic space, then matched against the audio.
Stage 04
Cue timestamps
Word boundaries become cue boundaries. Millisecond precision, zero transcription.
Why not just use a transcriber
Transcribers guess at your text.
We never do.
A speech-to-text tool listens and writes down what it thinks it heard. If your script already exists, that’s a step backwards.
What happens
Typical transcriber
With AlignSRT
Guest names & brand terms
Misheard, then mis-indexed by search
Exactly as you wrote them
Punctuation & line breaks
Generated, then manually fixed
Preserved from your source
After you re-cut the edit
Captions drift, you re-sync by hand
Re-run against the new audio
Cost per 100 minutes
$12–25 on most pay-as-you-go tools
$7. Credits never expire
Pricing
Pay for minutes.
Not for shelf space.
Subscribe monthly, or buy a pack once and keep it forever. Either way you are billed on audio length — never on video length, never on seat count.
Free
$0
First 30 minutes, fully downloadable
- No credit card needed
- Clean SRT, no watermark
- SRT & VTT export
Pay as you go
$3.99/ 100 min
One package, no expiry. Top up whenever you like.
- Credits never expire
- 100 minutes per pack
- Failed alignments are refunded
- Files up to 4 hours, no surcharge
Monthly
$9.90/ mo
500 minutes included · $1.98 per 100 min
- Batch upload
- Terminology lock (names, brands)
- Upgrade or cancel any time
Every month, pick your pool
Minutes refresh every month. Unused minutes do not roll over; buy a pack if you want a balance that never expires.
- Billed by audio length, not video length
- Unused pack credits stay yours, permanently
- Files deleted within 24 hours
FAQ
The questions people actually ask
Q1What if my script doesn’t exactly match what I said?
Small differences are fine — alignment handles filler words, stumbles and minor rephrasing. If a sentence was rewritten entirely, that line gets flagged and you can fix it in the editor. Every alignment reports a confidence score so you know exactly which lines to check.
Q2Do you count video length or audio length?
Audio length only. Upload a 100-minute video with a 100-minute voice track and it costs 100 minutes. We align the audio — video files are accepted but we don’t process the picture.
Q3What happens if the alignment fails?
Credits are returned automatically — not on request. If a file fails because of audio quality or a script that’s genuinely too different, you’ll see why and get your minutes back.
Q4How long can a single file be?
Up to 4 hours per file, with no surcharge for length. Audiobook chapters and full-length lectures are the reason we built it this way.
Q5Which formats can I export?
SRT and VTT on every plan, plus plain text. They import directly into Premiere Pro, DaVinci Resolve, Final Cut Pro and YouTube Studio.
Q6What do you do with my files?
They’re processed to produce your subtitles, then deleted within 24 hours. We don’t use your audio or scripts to train any model, and we never sell or share them.
Q7Can I get a refund?
If the output isn’t usable, email us within 14 days and we’ll refund the purchase — no questions about why. The free 30 minutes exists so you can judge the quality before paying.