Align
Word timestamps on device, for any transcriber. Cut, caption, and highlight on the word, five times more accurately, from 0.7MB.
Accurate word timestamps for any transcript.
Apple says "world" runs from 2.61 to 3.04 seconds. Align says 2.57 to 2.98. Cut there and the cut is clean, the caption lights up on the word, and the highlight starts where the speaker did.
You keep Apple's transcriber. Align reads the same audio SpeechAnalyzer already gets and returns the same words with tighter start and end times, in a few milliseconds, from a 0.7MB download. On clean audio that takes Apple's average miss from 113ms to 45ms.
A word Align can't improve keeps Apple's timing, so a correction can only help or leave the word alone. Nine languages, and a locale outside them passes through with Apple's timings untouched.
Five times more accurate.
Apple's word timings on LibriSpeech test-clean are off by 106.4ms on average; Align brings that to 20.2ms, and 95% of words land within 50ms, from a 0.7MB model.
Mean absolute distance from the reference word boundary, on a 500-clip sample of each official LibriSpeech split, with no speaker shared between training and evaluation
| Split | Apple raw | Align | Within 50ms |
|---|---|---|---|
| test-clean | 106.4ms | 20.2ms | 95% |
| test-other | 111.6ms | 24.8ms | 92% |
Evaluated on the v1.0.0 weights. The references are machine forced-alignment estimates from Qwen3-ForcedAligner averaged with a second aligner, not human annotations, so the figures show a large and consistent reduction of the timing error rather than sample-accurate ground truth. The one exception is the comparison against WhisperX, which is scored on 258 word boundaries a person corrected by hand against the waveform.
Use cases
Accurate word timestamps for any transcript.
Word-by-word captions that keep up
Light up each word as it is said in a captions view, a karaoke-style lyric screen, or a language-learning app, from Apple's transcript. The timings land on the word instead of a tenth of a second off it.
Cut a clip on a word
Trim a recording to a quote in a podcast editor or a video app and have the cut start where the word starts. No frame of the previous word, no clipped consonant, and no manual nudge on the timeline.
Highlight the spoken word in a transcript
Scroll a transcript in step with playback in a meeting recorder or a lecture app, with the highlight moving on the word rather than drifting ahead of the audio.
Search that lands on the moment
Jump to the second a word was said in an archive of recordings, on the device. The timings are tight enough that the playhead starts on the word and not in the middle of the previous one.
Inspiration
Ideas to build with Align. Copy a prompt into your coding agent and go.
Highlight each word as it's spoken, timed to the audio.
Run Apple's SpeechTranscriber with Desert Ant's Align attached as a timestamp refiner, then highlight each word using its refined time range. Align tightens Apple's word timings on-device; iOS 26 / macOS 26. Build it with the Desert Ant SDK. Align (Swift). Install and API: https://desertant.com/docs/align/. Install: SwiftPM desert-ant-core. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
Tighten a system transcriber's word timings so captions land on the word.
Attach Desert Ant's Align as a refiner to Apple's SpeechAnalyzer / SpeechTranscriber so its word-level timings tighten and captions and cuts land exactly on the word. iOS 26 / macOS 26, on-device. Build it with the Desert Ant SDK. Align (Swift). Install and API: https://desertant.com/docs/align/. Install: SwiftPM desert-ant-core. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
Build a text-based video editor where selecting words trims the clip.
Transcribe with Apple's SpeechTranscriber and Desert Ant's Align refiner for word-exact timings, then let the user select words to trim the video to that span. Deleting a sentence deletes the footage. iOS 26 / macOS 26. Build it with the Desert Ant SDK. Align (Swift). Install and API: https://desertant.com/docs/align/. Install: SwiftPM desert-ant-core. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
Add bouncing word-by-word captions to vertical videos.
Add animated word-by-word captions to short vertical videos, timed with Desert Ant's Align attached to Apple's SpeechTranscriber so each word pops exactly when it's said. On-device, iOS 26 / macOS 26. Build it with the Desert Ant SDK. Align (Swift). Install and API: https://desertant.com/docs/align/. Install: SwiftPM desert-ant-core. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
Make transcript search jump the audio to the exact word.
In an Apple media app, transcribe with SpeechTranscriber and Desert Ant's Align refiner so a search result seeks the audio or video to the precise word, not the sentence. iOS 26 / macOS 26. Build it with the Desert Ant SDK. Align (Swift). Install and API: https://desertant.com/docs/align/. Install: SwiftPM desert-ant-core. SDK source: https://github.com/Desert-Ant-Labs/desert-ant-core. Machine-readable catalog of every model and SDK: https://desertant.com/llms.txt.
What the model does
- Corrects the word-level timings Apple's
SpeechTranscriberandSpeechAnalyzerreturn, without replacing them: same words, same result surface, tighteraudioTimeRangevalues. - Mean timing error from 124.2ms to 43.9ms, macro-averaged across all nine languages so no single language carries the figure. On English it goes further: 106.4ms to 20.2ms.
- On a 500-clip sample of LibriSpeech test-clean, 95% of words land within 50ms of the reference, up from 37%. The worst tenth improves most: from 230.7ms to 33.0ms, about one frame of 30fps video.
- A structural fallback keeps Apple's original timestamp whenever a correction would be invalid, hits the search window's edge, or lacks streaming context.
- Nine languages: English, Spanish, French, Italian, Portuguese, German, Japanese, Korean, and Chinese. Other locales pass through unchanged.
- About 0.7MB of compiled Core ML in two stages, run on the CPU and Neural Engine; a typical result is refined in a few milliseconds.
Getting started
Add word timestamps to your iOS or macOS app in a few lines of code. Align docs.
// Swift Package Manager
.package(url: "https://github.com/Desert-Ant-Labs/desert-ant-core.git", from: "3.1.0")
// target dependency
.product(name: "Align", package: "desert-ant-core")import Align
let refiner = try await SpeechTimestampRefiner(locale: locale)
try await analyzer.start(inputSequence: inputs.recordingAudio(for: refiner))
for try await result in transcriber.results.refiningTimestamps(with: refiner) {
result.words // [WordTiming]: text, start, end, refined
}Add Align from Desert Ant Labs to this Swift project (iOS, macOS). What it does: On-device Word Timestamps for Any Transcript. SDK: Swift (iOS, macOS) Repo: https://github.com/Desert-Ant-Labs/desert-ant-core#readme // Swift Package Manager .package(url: "https://github.com/Desert-Ant-Labs/desert-ant-core.git", from: "3.1.0") // target dependency .product(name: "Align", package: "desert-ant-core") Reference: - Model page: https://desertant.com/models/align/ - Full catalog and other models: https://desertant.com/llms.txt Add the SDK, then follow its README for the exact API and current version. Do not invent API names or method signatures; confirm them against the README.
Specs
- Accuracy
- 20.2ms mean error on LibriSpeech test-clean against Apple's 106.4ms, and 43.9ms across all nine languages against Apple's 124.2ms
- On-device size
- About 0.7MB compiled Core ML (two 0.3MB stages plus the filter bank and calibrator)
- Languages
- English, Spanish, French, Italian, Portuguese, German, Japanese, Korean, Chinese
- Model
- Two-stage coarse-to-fine cascade over a log-mel spectrogram, 121k parameters per stage, with a gradient-boosted calibrator
- Platforms
- iOS 26, macOS 26, tvOS 26, visionOS 26 (Core ML), where SpeechAnalyzer lives
Align corrects a transcriber's timings; it doesn't transcribe, and the Swift SDK needs Apple's SpeechAnalyzer, which means iOS 26, macOS 26, tvOS 26, or visionOS 26. Align can't promise to improve every word.
The fallback that keeps the original timing catches corrections that look unsafe, not every wrong one. Spoken numbers are the weakest case: on a small sample, refinement moved digit boundaries further from the reference than leaving them alone.
FAQ
What is Align?
Word timestamps on device, for any transcriber. Cut, caption, and highlight on the word, five times more accurately, from 0.7MB.
Does Align run on device?
Yes. Align runs on the device, with no server call, so the data stays with the user.
Which platforms does Align support?
Align ships as a native on-device SDK for Swift.
How much does Align cost?
Every model is free up to 100k monthly active devices per SDK. Unlimited inference per user. Contact us for custom licenses.
How accurate or fast is Align?
Apple's word timings on LibriSpeech test-clean are off by 106.4ms on average; Align brings that to 20.2ms, and 95% of words land within 50ms, from a 0.7MB model.