Premove ITN

Turn spoken ASR text into written values such as phone numbers, dates, times, amounts, email addresses, URLs, and identifiers. Premove ITN uses sentence context to choose between valid interpretations.

the room code is one oh five
→ the room code is 105

In the frozen evaluation, the same spoken value could be mistaken for a time. Premove selected the room code in this example.

Why voice agents need ITN

Speech recognition returns text in the form people say it. A tool often needs a structured value instead: seven eight three two nine becomes an order ID, and twenty dollars becomes an amount. A formatting mistake can become a failed lookup or an invalid tool argument. Learn where ITN fits in a voice-agent pipeline.

How Premove works

Generate → Score → Decode

Rust generates valid written candidates. A frozen DeBERTa scorer uses the full sentence to rank them. An exact decoder selects compatible edits and can leave text unchanged when an edit is not justified. The model cannot output a form that the deterministic candidate layer did not generate. Read the illustrated explanation.

What it can write

Spoken form	Possible written form
`one hundred twenty`	`120`
`four thirty`	`04:30`
`twenty dollars`	`$20`
`support at example dot com`	`support@example.com`
`v three dot one dot nine`	`v3.1.9`

These are supported candidate forms, not a promise that every sentence will select that candidate. Context determines the final output. See supported forms and boundaries.

Measured results

Premove ITN scored 398/400 (99.50%) on the voice-agent subset of its frozen synthetic stress benchmark. Its overall semantic entity accuracy was 89.70% across the 1,500-row suite. Mean warm latency was 56.49 ms on the measured Apple M4/MPS setup. The benchmark is not a production-traffic accuracy estimate; Premove was also slower than both comparison backends. See the full comparison, method, and limitations.

Install and try it

pip install premove-itn

from premove_itn import PremoveITN

itn = PremoveITN.from_pretrained()
print(itn.normalize("the room code is one oh five"))
# the room code is 105

The first use downloads about 1.6 GB of model files. Load one instance and reuse it for later requests. The frozen-model release is validated on Apple Silicon MPS and Linux x86-64 CPU; see deployment for the complete platform boundary.

Start here

Integrate

Understand

Run in production

Learn

Internals

Why voice agents need ITN

How Premove works

What it can write

Measured results

Install and try it

Open source and open weights

​Why voice agents need ITN

​How Premove works

​What it can write

​Measured results

​Install and try it

​Open source and open weights

Why voice agents need ITN

How Premove works

What it can write

Measured results

Install and try it

Open source and open weights