Gladia Solaria is launched as the Multi -Language Language Forms based on the transcription of speech to the text

Gladia, an AI’s intelligence provider and sound intelligence, has launched Solaria, which is the ASR recognition model from the next generation designed to redefine communications in the actual time of communication centers and other first audio platforms.

Solaria now enables companies to enhance and expand their customer service operations through sound technology on behalf of artificial intelligence that provide unparalleled linguistic coverage-which supports more than 40 languages ​​previously unknown with other solutions-with quality or speed solutions.

Although the use of external sources was a strategy to long reduce costs in the communication center industry, companies are now facing a new and decisive challenge: providing widely multi -language smooth support. With 49 % of the international executives who reported financial losses due to language barriers, the demand for multi -language solutions was not larger and high -quality.

“We have seen a significant increase in the market in the AI,” said Jean -Luis Kogener, CEO of GMESBEAT.

He said the product also contains features such as morale analysis in actual time and real time translation. He deals with speech with the translation of the text and copying. It is important to do it in real time for audio agents or communication centers, as someone may have to answer a question that comes in a different language.

Solaria: A ready -made model for institutions for the experience of global customers

Solaria is a STT engine designed for global expansion. Solaria is designed to meet the requirements of communication centers today, where both artificial intelligence automation and human agents need high support, low word, and multi -language support in the actual time of success.

The model achieves pioneering results in industry in speech recognition, providing both accuracy and rapid treatment. Recent criteria show that Solaria has reached the average rate of words (war) by 94 % in English, Spanish, French and other common languages, while maintaining a low transition time of 270 milliliters, which makes the conversation feel natural and response.

While talking to the text in the actual time is often measured at speed alone, the accuracy and linguistic coverage are two decisive things with the same amount for companies that provide smooth services across the regions.

Unlike speech models to other text that give priority to speed on the ability to use, Solaria is a balance between the leading accuracy in industry and speed with unparalleled linguistic coverage-100 languages ​​in total, with exclusive support for 42 languages ​​with which competitors do not match. For high -population markets and corridors of external sources such as Bangladesh, India and the Philippines, accuracy of the original level is now provided in regional languages ​​through Solaria.

With copying at the original level, switching code in actual time, translation across all supported languages, companies can expand in global markets without restrictions.

Solaria is designed for the organization’s vocal automation,

The best accuracy in its class in high -population languages Such as Tagalog, Bengali, Punjabi, Tamil, Urdu, Persian, and Marathy.

The ability to adapt the form To the terminology of the industry (such as medical or financial terms) and make them extract important data, such as names, addresses and number.

Adaptive speech treatmentEnsure high accuracy in the loud contact center environments.

Data safety at the institution levelIn full compliance with GDRP, HipPa and Soccer 2.

With the addition of Solaria to its product portfolio, GLADIA allows companies to enhance customer service by improving audio agents who work with artificial intelligence, making IVRS and apparent aids more reliable across multiple languages, with improving work flows that help human beings with copies and translations in actual time to help provide more effective assistance.

“Speech is the most natural way to communicate with the world-for the first time, that automated identification of speech is the closure of the gap, which enables humans and Amnesty International to speak the same language really,” Jean-Louis Qio, CEO of Gladia, said in a statement. “With Solaria, we have achieved a breakthrough in the technology of Voice operating in Amnesty International that opens new opportunities for companies, leading efficiency and providing more unpleasant and effective customer experiences across languages ​​and various markets. Solaria is designed for sound platforms from the next generation ready to lead this global transformation.”

It serves more than 700 institutions clients around the world, including attention, Circleback, Method Financial, and VEED.O, Gladia provides service at the level of institutions and expansion, supported by support and infrastructure in the United States and Europe, ensuring reliable performance of critical applications. Companies looking to expand their scope can improve operational costs, and enhance customer experiences construction using API’s API today.

As part of the launch of Solaria, Gladia has made a partnership with Livekit, a pioneering framework on open source developers for the actual time of artificial intelligence agents, to the actual time translation within the AI’s applications. This gives developers the capabilities of global language outside the box through smooth integration with API’s Gladia.

In the aftermath of a round of $ 16 million in 2024, Gladia was launched another critical step towards establishing itself as a pioneering provider in API from one side to party-learn about speech recognition, and the Toyodic AI, and sound generation capabilities to help the institutions and developers of the full potential of vocal data at the time Actual.

Paris-based Gladia was founded in 2022 by Jean-Louis Queguiner (EX-FHCLOUD) and Jonathan Soto (Ex-Mit/Sigfox). The Gladia product has been adopted by more than 150,000 users and 700 clients from the institution – including industrial leaders such as attention, Circleback, Method Financial, and VED.IO.

There is a delay of 300 million MLICEN between the moment when the speaking begins and the moment when you receive the first event to activate the sound. It takes 100 milliseconds to copy and thus you have immediate results.

To improve accuracy, Queguiner said the company needs to train in more data. It needs to work while increasing data to make data more powerful. The company has the prices of institutions in the price, but it has not yet been revealed. He said he would be among the most reasonable solutions on the market.

The company has approximately 40 employees.