Verified, then logged.
Sun 22 Nov 2026 · 00:21 KST / 15:21 UTC한국어 | ENSearch /

02 / IT · Analysis

DeepL unveils Aure to track changes in spoken expression

DeepL described a move from preserving a speaker’s voice to tracking changes in delivery, but its Korean product page still labels voice-to-voice translation as coming soon.

An illustrative meeting scene shows a speaker gesturing toward a microphone, a listener wearing an earpiece, and paired audio waveforms reflecting changes in speech intensity and rhythm. AI-generated image

DeepL announced on 11 Nov that Aure, a model designed to reflect changes in tone, emotion, emphasis and pace in translated speech, would apply automatically for DeepL Voice users.

The company described a shift from preserving a recognisable voice to following how that voice changes during a conversation. That distinction underpins the update, while the Korean product page raises a separate issue about availability.

DeepL lists Korean among Aure’s initially supported languages. Its Korean product page says pricing depends on capacity, product type and usage, with customer-specific quotes available through the sales team.

From voice preservation to changing delivery

DeepL explained that its earlier approach used a voice sample from the opening moments of a conversation to determine the translated voice. It preserved the speaker’s vocal identity but could not follow subsequent changes in emotion, emphasis or pace.

According to the Aure announcement, the new model uses both source audio and translated words to shape sentence delivery. The stated change concerns how someone speaks as well as the words being translated.

In its earlier announcement, DeepL said voice preservation covered 14 languages. The Aure announcement says existing voice models continue to serve languages that Aure does not support.

Korean guidance still signals a pending launch

DeepL previously announced general availability of voice-to-voice translation for online meetings. Its Korean product page still describes the same feature as coming soon, leaving a difference between the official announcement and the local product guidance.

The company said Aure applies automatically across online meetings, in-person conversations and its voice API, a developer interface for connecting voice translation to other services. No additional configuration is required.

Korean language support and the timing of access are separate matters. DeepL has not disclosed why its Korean product page retains pending-launch wording after the general-availability announcement; that discrepancy leaves the timing of access unresolved in its Korean guidance.