BERTIN-IT
5.6.2019 10:17:03 CEST | Business Wire | Press release
Bertin IT (CNIM Group) announces the release of the new version of MediaSpeech®, its multilingual speech recognition solution that converts audio tracks to searchable text transcripts, enabling audio and video sources, to be indexed searched and analysed. MediaSpeech® now also comes in a live version for real-time audio streams, paving the way for new interactive and augmented communications applications.
This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20190605005384/en/
Thanks to deep neural networks commonly used in Artificial Intelligence systems, MediaSpeech® creates an extremely fine model of the acoustic space which is robust with different speakers and acoustic conditions, so offering even faster and more accurate transcription.
Features:
- Speech recognition with each word being transcribed within a millisecond and assigned a recognition confidence score .
- Automatic detection of spoken language (LID).
- Automatic segmentation speaking slots and speakers with gender recognition.
- Identification of the speaker from a biometric database.
- Automatic and semi-automatic adaptation of vocabularies and domains.
And all this in 17 different languages.
MediaSpeech® has several variations: deployed on site or in SaaS mode , hosted on Bertin IT’s cloud, MediaSpeech® Factory can handle large volumes of files with guaranteed performance levels ; a new version MediaSpeech® Live is able to transcribe audio streams on the fly, opening the door to innovative real-time applications - voice chatbots, call-bots, enhanced call centres (the enhanced call centre concept involves the provision of assistance to the adviser during the call so streamlining and improving the quality of the dialogue.).
Among the main improvements in the new version of MediaSpeech®:
- MediaSpeech® Live version for processing audio streams in real time.
- New neural models make transcription two to three times faster and more accurate.
- “Full" neuronal transition of all speech processing modules : speech detection (VAD) and speaker segmentation (Diarization) for even greater accuracy.
- Easy installation process , stronger security and new interfaces.
- A fully neuronal language identification module (LID) with increased accuracy , even for relatively short sections of speech.
Version 6 of MediaSpeech® is already being used by several customers, including a major French investment and finance bank. The MediaSpeech Live version has just been delivered to another major banking group for use at its contact centres.
View source version on businesswire.com: https://www.businesswire.com/news/home/20190605005384/en/
Contact:
Nathalie Sablon nathalie.sablon@cnim.com
Link:
About Business Wire
Subscribe to releases from Business Wire
Subscribe to all the latest releases from Business Wire by registering your e-mail address below. You can unsubscribe at any time.
Latest releases from Business Wire
Beacon Acquires Haize Labs to Power AI for the Real Economy17.9.2026 16:30:00 CEST | Press release
Haize Labs leadership will anchor Beacon’s new applied AI research group, accelerating the deployment of AI across the essential businesses Beacon owns Beacon Software (Beacon), the first centralized AI holding company, today announced that it has acquired Haize Labs, an AI reliability company that tests and safeguards AI agents for real-world deployment. The Haize Labs team will join Beacon to form its Applied AI Research Group, with co-founder and CEO Leonard Tang joining as VP of AI Research to lead the AI operating system that is deployed across Beacon’s companies. Beacon believes that the fastest way to bring AI to the traditional economy is through the software businesses that essential industries already use and trust, equipping them with AI infrastructure. Beacon acquires these businesses for long-term growth and connects them to a centralized AI operating system that compounds with each additional acquisition and industry. The addition of the Haize Labs team will strengthen th
TIMELAB Develops OEM/ODM Hair Straightener Reaching 200°C at All Three Measured Points in 15 Seconds17.9.2026 16:00:00 CEST | Press release
Testing at 230 V with a 260°C setting confirmed at least 200°C after 15 seconds at all three measured plate locations (front, center and rear), not just a single point. TIMELAB Co., Ltd., a Japanese beauty-appliance OEM/ODM provider, has developed a hair straightener for OEM/ODM customization for international hair care brands and professional markets. In testing at 230 V with a 260°C setting, all three measured plate locations reached at least 200°C after 15 seconds. This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20260914187865/en/ Three-Point Plate Testing A European-market development unit with a maximum setting of 260°C was tested at 230 V. Fifteen seconds after power-on, readings at the front, center and rear were 212.5°C, 227.5°C and 200.0°C, respectively. From around 60 seconds onward, temperature differences among the three points narrowed, indicating a stable trend. Precision Temperature Control An NTC sensor inside t
Earnix Brings Agentic AI to the Decisions That Drive Insurance Performance17.9.2026 15:09:00 CEST | Press release
Agent Hub, part of Earnix AIOS, brings more than 25, insurance-specific agents and apps into Earnix’s governed pricing, underwriting, and customer decision solutions- improving workflows and business performance Earnix today announced the introduction of Agent Hub, a curated catalog of insurance-specific AI agents and apps within Earnix AIOS — the AI Orchestration System powering Earnix’s pricing and rating, underwriting, and customer engagement solutions. Agent Hub brings agentic AI directly into these solutions, enabling more intelligent workflows and strengthening the high-stakes decisions that shape pricing, underwriting, customer engagement, growth, and profitability. In an increasingly turbulent insurance environment, the useful life of an insurance decision is getting shorter. Risk and market conditions are changing faster, making growth, profitability, and portfolio performance harder to manage. AI is increasing the speed and scope of what insurers can analyze, recommend, and i
Medisca Launches Global Peptide Division17.9.2026 15:07:00 CEST | Press release
New division brings API-grade peptides and Medisca’s quality-first pharmaceutical expertise to a rapidly evolving area of personalized medicine Medisca today announced the launch of Medisca Peptides, a dedicated global division focused on expanding access to API-grade peptides and building an integrated peptide platform for pharmaceutical and personalized medicine markets worldwide. As the peptide landscape evolves, Medisca is proactively building the capabilities and infrastructure needed to support API-grade peptides within applicable regulatory frameworks. The launch represents a significant long-term commitment to peptides, including the expansion of Medisca’s portfolio, strategic manufacturing relationships, and the pharmaceutical infrastructure needed to support the category’s growth. It also signals Medisca’s ambition to become a leading force in the next generation of peptide innovation. For Medisca, the opportunity is familiar. More than 35 years ago, the company was founded t
Elliptic Launches Its Decode Agent to Answer On-Chain Risk Questions and Shows Its Work17.9.2026 15:00:00 CEST | Press release
Hours of tracing, answered in seconds, with the evidence attached Elliptic, the global leader in on-chain risk, today launched Decode, an AI agent built on Elliptic’s intelligence platform. The Decode agent answers natural-language questions about cryptoasset addresses and their exposure in seconds, and returns the evidence behind every answer. Compliance and investigations teams need answers to questions nobody planned for. Getting them out of the data is now a specialist skill, so the questions wait for someone who has it, or go unasked. Meanwhile illicit actors move at machine speed, and agents are starting to sit on both sides of a payment. General-purpose AI assistants can produce an answer. They cannot show where it came from. A fluent response with nothing behind it cannot be put in front of a second-line reviewer, an auditor, or a regulator. “The market is full of AI that asks to be trusted. That is not good enough when someone asks how you reached a decision,” said Jackson Hul
In our pressroom you can read all our latest releases, find our press contacts, images, documents and other relevant information about us.
Visit our pressroom