CEVA, Inc., licensor of wireless connectivity and smart sensing technologies, and Fluent.ai, a provider of on-device, small footprint and multilingual speech understanding solutions, announced that the companies have partnered to offer ultra-low power speech-to-intent solutions for intelligent edge devices. Fluent.ai’s suite of speech-to-intent technologies has been ported and optimized for CEVA’s low power audio and sensor hub DSPs, providing a high performance solution for OEMs and ODMs looking to integrate intelligent voice activation and control into their wearables, consumer devices and IoT products. Fluent.ai provides embedded, noise robust and multilingual speech understanding solutions capable of running offline on small footprint and low power devices. Fluent.ai technology can support any language and accent, enabling users to speak to their devices in their native language, naturally, and without sacrificing their privacy. CEVA’s audio and sensor hub DSPs, including the CEVA-X2, CEVA-BX1, CEVA-BX2 and SensPro family, enable the full suite of speech-to-intent technologies to run in always-on mode.
Category: Computing & data (Page 57 of 80)
Computing and data is a broad category. Our coverage of computing is largely limited to software, and we are mostly focused on unstructured data, semi-structured data, or mixed data that includes structured data.
Topics include computing platforms, analytics, data science, data modeling, database technologies, machine learning / AI, Internet of Things (IoT), blockchain, augmented reality, bots, programming languages, natural language processing applications such as machine translation, and knowledge graphs.
Related categories: Semantic technologies, Web technologies & information standards, and Internet and platforms.
Qlik announced enhancements to Insight Advisor, its AI assistant built into Qlik Sense, to deliver augmented intelligence capabilities for cloud analytics. Drawing on Qlik’s Associative Engine, combined with natural language processing (NLP) and cognitive technology. Users can interact with Insight Advisor in a variety of different ways, including search-based visual analysis (NLP-driven), conversational analytics (chat), associative insights to expose hidden data relationships, assistance with creation and data preparation, and advanced analytics calculation and integration. Updated features include:
- Insight Advisor Chat – A new, fully conversational analytics experience native to Qlik Sense SaaS, available in the Qlik Sense hub in multiple languages. Uses NLP and natural language generation to understand user intent and generate both narrative and visual responses.
- Business Logic – Ability to create business rules and metadata to customize and guide Insight Advisor’s behavior when generating insights and understanding natural language.
- Advanced Analytics Calculation – A new function, K-Means Clustering, allows data points to be grouped together based on similarity and is highly useful for customer segmentation, fraud detection and many other use cases. In addition, Insight Advisor will now auto-generate cluster and correlation charts in search-based visual analysis.
- Search-based Visual Analysis on Mobile – In addition to Chat, search-based insight generation is now available on handheld devices.
DataStax announced the general availability of Storage-Attached Indexing (SAI) for Apache Cassandra available on Astra and DataStax Enterprise (DSE). DataStax has also opened a Cassandra Enhancement Proposal (CEP) with the Apache Cassandra project to share this with the open source community. With Storage-Attached Indexing, developers now have accessibility to familiar indexing and queries – such as WHERE clauses – in Apache Cassandra. Storage-Attached Indexing is an index implementation that enables users to index multiple columns on the same table without scaling issues. The benefits of Storage-Attached Indexing include:
- Improved stability
- Significantly reduces disk usage
- Improved numeric range performance
- Releases constraints to data modeling and barriers to scale-out
- Features modern and expected indexing features on Apache Cassandra
Who benefits from schema. org?
Schema.org, linked data, and knowledge graphs are powerful tools for organizing and navigating vast amounts of information. Much of the current energy around these tools is related to SEO and search engines, especially Google, who depend on them to provide a better search experience. These same tools help commercial and corporate publishers deliver better, and more unique, web experiences to researchers and other content consumers.
We all have a stake in how well these tools work, so we need to understand the process of creating and managing them, and how stakeholders share the cost, risk, and benefit of the raw material, technical development, and maintenance.
Content strategist Michael Andrews’ deep dive into the history and process behind schema.org’s management is an enlightening read for stakeholders.
Taming the tail: adventures in improving AI economics
Martin Casado and Matt Bornstein focus on the business models and challenges of machine learning companies and products, which are more unique than you might realize and something we need to learn a lot more about. We recommended an earlier article of theirs on the differences between the business models of AI companies and software companies. This article is a follow-up and provides some guidance on how to deal with some of the challenges previously identified. Especially interesting is their example of long-tailed distributions to illustrate the importance problem understanding.
Headless meets serverless – a tierless architecture for frictionless enterprise
The components of modern enterprise IT architectures have changed considerably in the last few years. The use of APIs, microservices, XaaS (everything as a service), headless, and serverless approaches have, individually and especially in conjunction, become strategically critical. As Phil Wainewright puts it…
As these connected digital technologies mesh together, they begin to reshape the nature of the enterprise, opening up new ways to collaborate, connect and do business. We are still at the very beginning of adjusting to what this means for how we live and work.
Wainewright explains what these technologies are, describes related activity and trends, and makes a case for a tierless model. His article is relevant and will be helpful to both IT and business managers.
The field of natural language processing is chasing the wrong goal
Researchers are too focused on whether AI systems can ace tests of dubious value. They should be testing whether systems grasp how the world works.
Also…
- Another large-scale knowledge graph… This know-it-all AI learns by reading the entire web nonstop via MIT Technology Review
- What Could Go Wrong? Opportunities and pitfalls for multi-user AR experiences via ARPost
- On the waitlist for this. A plan to build a search engine Google can’t beat via Protocol
- Bye Bye PhoneGap, hello PWAs. Update for customers using PhoneGap via Adobe I/O
The Gilbane Advisor is curated by Frank Gilbane for content technology, computing, and digital experience professionals. The focus is on strategic technologies. We publish more or less twice a month except for August and December. We do not sell or share personal data.
Subscribe | Feed | View online | Editorial policy | Privacy policy
Photonic quantum computing company, Xanadu, announced the release of the first publicly available photonic quantum cloud platform. Developers can now access Xanadu’s gate-based photonic quantum processors, in 8, 12, and soon 24 qubit machines.
Photonics based quantum computers have advantages over older platforms. Xanadu’s quantum processors operate at room temperature. They can easily integrate into existing fiber optic-based telecommunication infrastructure, enabling a future where quantum computers are networked. It also offers great scalability supporting fault tolerance, owing to robust error-resistant physical qubits and flexibility in designing error correction codes. Xanadu’s unique type of qubit is based on squeezed states – a special type of light generated by their own chip-integrated silicon photonic devices.
Xanadu’s partners and customers are currently testing solutions on the pre-release Xanadu Quantum Cloud. These include academic institutions, quantum startups, and national labs including Creative Destruction Labs, Scotia Bank, BMO and Oak Ridge National Laboratory (ORNL). Now the company will be extending access to a number of new enterprise clients who are seeking to leverage quantum computing to solve their most complex problems.
In addition to Xanadu Quantum Cloud, developers will use Xanadu’s open-source tools widely available on Github. These include Strawberry Fields, its cross-platform Python library for simulating and executing programs on quantum photonic hardware and PennyLane, its software library for quantum machine learning, quantum computing, and quantum chemistry. The community includes a vast array of tutorials and educational materials for users of all levels of experience to begin developing and experimenting with quantum applications.
The enterprise adoption of quantum computing is in the early stages of development, but access to photonic quantum computing over the cloud will give developers across industries and academia the chance to explore potential business applications. Open access educates and drives interest towards new, concrete implementations that demonstrate the future of computing. Xanadu’s processors provide researchers and developers with novel approaches that are unique to solve problems in finance, quantum chemistry, machine learning, and graph analytics.
Computing based on quantum mechanics. It promises to dramatically change computing power, uses, and perhaps the political balance power. It is not quite ready for widespread enterprise or research use but it is coming (see Xanadu releases Quantum Cloud platform) and companies and countries are accelerating investment in efforts to be ahead of the curve.
Where traditional computing is based on bits representing either 1 or 0, basic particles in quantum mechanics, “quantum bits”, or “qubits” can take on a third state that is not 0 or 1 but one of any probable number of undefined in-between states before being detected. This undefined state is referred to as “superposition”.
These superpositions can be “entangled” with superpositions of other objects no matter the distance. This is at least part of why Einstein was uncomfortable with quantum mechanics’ threat to his special theory of relativity and what he called “spooky action at a distance”. Entanglement may seem spooky but it has been demonstrated plenty of times.
Postman announced the launch of Postman for the web, a new browser interface for Postman that offers users simplified access and superior collaboration. The move marks somewhat of a return to Postman’s original roots. Until today, Postman users have accessed Postman through a desktop application. Accessing Postman through the browser has been one of the most-requested features from users; they say that downloading a desktop app can slow down onboarding, and can also make it challenging to stay on the latest version. Plus, the browser approach allows certain collaboration capabilities that aren’t possible within the app. The web version of Postman solves these issues by providing the following:
- Simplified user access and onboarding: Users get instant access via browser with no app installation required, automatic updates, and faster user onboarding (i.e., just paste in the URL and go).
- Optimized collaboration through deep linking: Now that Postman is on the web, everything in Postman has a URL, so users can easily share URLs to granular API elements for superior collaboration.
- The ability to send requests at scale from the browser: A unique agent architecture, developed exclusively by Postman engineers, facilitates scalable sending of API requests from the browser.
Tightly connected to the Postman for the web client launch is the Postman agent. The agent was specifically designed to overcome the cross-origin resource sharing (CORS) limitations of browsers to allow for API requests to be scalably sent from a browser interface. By creating a proxy micro-application to facilitate API request-sending at scale, the agent addresses this browser challenge in a way that’s never been done before.
ThoughtSpot launched ThoughtSpot 6.2 to help enterprises connect, share, and utilize insights faster. New features include DataFlow, Embrace for SAP and Teradata, and improvements in user experience, data exploration, and data loading. ThoughtSpot’s search and AI-driven analytics platform was designed to unlock insights for employees at every level of an organization, from frontline employee to C-suite executive. Features include:
- Answer Explorer 2.0: Leverages AI to guide users to unasked questions and new insights for which they would not have thought to look, automatically curating suggestions, recommended additional searches, and unexpected anomalies or trends insights directly in a pinboards, charts, and answers. Built in machine learning continuously improves Answer Explorer, tailoring insights for users.
- DataFlow: DataFlow makes it easier to bring data into Falcon, ThoughtSpot’s in-memory database, through a no code point and click UX, enabling enterprises to utilize search and AI-driven analytics even for data sources too slow to run search queries directly.
- Embrace Expansions: ThoughtSpot Embrace is now available for both Teradata and SAP HANA, allows users to run search and AI-driven analytic inquiries directly in these databases without the need to move or cache any data. Improved user experience makes it to connect Embrace directly to Snowflake, Amazon Redshift, Google BigQuery, Microsoft Azure Synapse, SAP HANA, and Teradata.
- API-Powered Bulk Load Connector: A new API-based feature enables the smooth and secure transfer of data to Falcon, ThoughtSpot’s in-memory database, by allowing users to load data in bulk from ETL tools and custom programs.
- Caffeine for Advanced Data Caching: Caffeine gives users the freedom to choose which pinboards are pre-cached and when, allowing them to optimize the performance of the most important pinboards.
- Scriptability Support for Worksheet Filters: Worksheets with filters can be exported and imported to the same environment or between environments.
- Improved Visualizations: ThoughtSpot’s high cardinality charts now accommodate up to 35,000 data values, an increase from 1,000 previously. The high cardinality chart legend limit has also increased from 40 items to 250 items, giving pinboard creators the ability to display much richer data for their users.
- Data Compression: New in-memory data compression makes it possible to leverage more data without increasing storage, reducing costs as organizations scale their data footprint.
Current ThoughtSpot customers will have access to ThoughtSpot 6.2 today.
Subscribe to the Gilbane Advisor
