AIMPACT News, May 20 (UTC+8), according to Beating monitoring, simultaneous interpretation systems are evolving from monotonous voice translation to full-modal digital interpreters that can understand visuals and clone human voices. On May 19, Alibaba Tongyi Lab officially announced the launch of a new generation of real-time audio and video simultaneous interpretation model, Qwen3.5-LiveTranslate, which significantly upgrades real-time simultaneous interpretation capabilities to over 3,500 language pairs, and for the first time supports real-time voice cloning, hotword customization, and visual understanding. The new model, based on the Qwen3.5-Omni architecture, now supports understanding and writing in 60 languages, as well as voice output in 29 languages. Unlike traditional simultaneous interpretation software that only listens to audio, the new model introduces real-time visual context to eliminate semantic ambiguity. For example, when a specific type of mask appears in a video frame, the system can combine visual features to accurately distinguish between a medical mask and a masquerade ball mask in English, thereby compensating for the lack of audio information. To eliminate transcription bias caused by noise and accents, the new model also introduces a dynamic hotword injection mechanism. The reason is simple: users can directly specify specific names, brands, or industry terms in the translation stream, forcibly locking in the correct translation and preventing proper nouns from drifting during simultaneous interpretation. During cross-language simultaneous interpretation, the model also supports real-time voice cloning, enabling real-time reproduction of the speaker's original voice timbre and tone in the interpretation stream. Currently, the new model is open on the experience platform Qwen Omni, and the API will be launched on the Alibaba Cloud Bailian platform in the future. (Source: BlockBeats)

View Original

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.

Reward
like
Comment
Repost
Share

Comment

Add a comment

No comments

Trending Topics
View More
#
Get2SharesOfSKHynixAtZeroCost
1.51M Popularity
#
MicronOvertakesMetaInMarketValue
33.04K Popularity
#
WorldCup🇫🇷vs🇳🇴
316.97K Popularity
#
USMayPCEInflationRisesTo4.1%HighestIn3Years
521.02K Popularity
#
StakeUSD1Earn9.48%APR
978.51K Popularity

Pinned

Sitemap

Supports 3500 language pairs! Alibaba releases the first visually enhanced simultaneous interpretation large model Qwen3.5-LiveTranslate

Trending Topics

Get2SharesOfSKHynixAtZeroCost

MicronOvertakesMetaInMarketValue

WorldCup🇫🇷vs🇳🇴

USMayPCEInflationRisesTo4.1%HighestIn3Years

StakeUSD1Earn9.48%APR

Pinned