Space- Efficient representation of entity centric query language models

September 19, 2022

Automatic speech recognition (ASR) is used by virtual assistants to aid users in answering entity-centric enquiries. However, because of the vast quantity of continuously changing named things, spoken entity recognition is a tough task. Furthermore, when ASR is conducted on-device, the resources available for recognition are limited. 

We investigate the use of probabilistic grammars as language models within the finite-state transducer (FST) framework in this blog, where a deterministic approximation to probabilistic grammars is introduced that avoids the explicit expansion of non-terminals at model creation time, integrates directly with the FST framework, and is complementary to n-gram models.

In this blog, we’ll also look at how entity-centric probabilistic grammars may be used as language models (LMs) in Automated Speech Recognition (ASR) to improve the recognition of entity-oriented inquiries, which are widespread in VA use. 


A generic language model trained on transcribed VA questions is supplemented by an entity-centric language model of spoken VA queries. The purpose is to improve voice recognition of tail entities while keeping resource restrictions in mind (i.e., model size). ASR language models give a non-zero probability P (wk |w1,...,wk-1) to words wk that are preceded by left-context w1,...,wk-1, with words wi belonging to a fixed-size vocabulary V. The language model probability is coupled with auditory information during ASR decoding to distinguish between competing hypotheses. 

First, the framework involves showing how entity-oriented inquiries may be represented as probabilistic grammar. Following that, non-deterministic graph grammar representations and their drawbacks and finally, the method, φ-RTN, offers a close approximation to FST-compatible graph-based grammars.

Entity-centric Query Grammars

An investigation of a month's worth of anonymised English query logs from a popular VA reveals that more than 15% of media player requests instructing the VA to play a song, album, or artist use the "play entity" form. As a result, a large proportion of media player queries can be represented using probabilistic grammar. The advantage of estimating a grammar using entity-centric queries derived from use logs—which will then be utilised for language model estimation—is that we may maintain the query templates unchanged while updating the weighted entity list using external knowledge sources that correspond with usage. As a result, the VA may distinguish new items that are not yet present in transcribed training data. 

Denote T and E as the sets of query templates and entities, respectively, and show an example grammar G (composed of templates T with slots for entities E) that we want to support. Domain experts reviewed the templates derived from use logs, while the entity feed was extracted from an external knowledge source. We are particularly interested in increasing language model coverage for searches including tail entities. Tail inquiries are those whose combined probability is less than the median.

Recursive Transitive Networks

Encoding the grammar shown as a static FST by expanding every occurrence of the entity in the templates results in models that are proportionate to the cross product, |T||E|. Recursive Transitive Networks (RTNs) provide an alternate form for encoding grammar and, more broadly, class-based language models, which do not store the cross-product. 

An RTN, on the other hand, is made up of a family of FSTs, each of which is connected with a non-terminal, where non-terminal labels on arcs inside the component FSTs are recursively replaced by the FST they point to. The non-terminal symbol S denotes the root non-terminal of the constituent FST that is investigated first. Grammar G may be expressed by an RTN with a single degree of recursion.

Unfortunately, RTNs are often non-deterministic when employed as decoding graphs in an FST decoder. If there are many outgoing arcs that share the same input symbol at a given state, the FST is nondeterministic. RTNs exhibit non-determinism since various pathways across the graph results in the same token sequence. Inside the grammar, this would be the case within the state corresponding to context "hello VA"—that is, the wake word supplied by templates—because the following token "play" can match either template or entity. Non-determinism is harmful to effective ASR decoding since numerous pathways might lead to the same hypothesis and should thus be avoided.

RTNs with Deterministic Approximation

We will now describe the topology of our model, dubbed φ-RTN, as a two-level RTN, and explain how our model is made deterministic (approximating the actual RTN) by applying precedence rules and -transitions. Furthermore, we describe how we guarantee that our model is properly normalized.

Topology: A two-level RTN is used, with the root nonterminal S linked to a rigid grammar FST FT calculated on templates. To account for out-of-domain utterances, the transition probabilities for each stage are doubled by discounting factor α = 1 − α (0 < α < 1). Template FST FT makes use of a non-terminal entity, which is related to entity FST FE. FST FE is represented by a standard, non-backoff n-gram model over entity names, with probabilities scaled to accommodate out-of-domain entities. The remaining mass at each FE state will be utilised to transition out of FE, back to FT, and maybe sP (w) in the event of out-of-domain entities.

Determinism: We address the previously reported RTN determinization issue by favouring regular symbols over nonterminal entry/exit. Within FT, observing non-terminal entity after context w1,..., wi, with associated state s, the appropriate sub-FST FE may be accessed by following the φ-transition from state s. It should be noted that in the absence of an entity, the -transition results in the unigram state sP. We remember the state we need to return to in FT when we enter FE. φ-transitions are used within FE to depart sub-FST FE and return to template FST FT. Final states in FE are deleted, and instead, the FST state's final probability mass is merged with the residual mass and divided among normal symbols at each state s in FE.

Normalization: We use an approach similar to back-off n-gram models to verify that the final model is suitably normalized for all states s. More precisely, we add a weight to the φ-transition leaving state s such that the probability mass attributed to unseen events at s is scaled to fit within the residual mass at state s when following φ-transitions recursively. Because we know the initial state in FE that the φ-transition leads to, and the state from FE that leads back to FT after following once more, the weights for φ-transitions leading from FT to FE can be computed statically when the model is built.

However, because of the reliance between the state in FE from which we are quitting and the state in FT to which we are returning, computing the weights for φ-transitions that exit from FE to FT (excluding the start state in FE) is more complex. More precisely, we add a weight to the φ-transition leaving state s such that the probability mass attributed to unseen events at s is scaled to fit within the residual mass at state s when following φ-transitions recursively. Because we know the initial state in FE that the φ-transition leads to, and the state from FE that leads back to FT after following once more, the weights for φ-transitions leading from FT to FE can be computed statically when the model is built.

At runtime, because we know the residual mass at state s in FE = α + P (final|s), we only need to evaluate explicit events specified at state s in FT that result from the φ-transition exiting FE to update the partial weight. Because FT is sparse in our case, the computation associated with this process is insignificant. It should be noted that φ-RTN does not accept templates with two or more consecutive non-terminals since doing so would result in much more processing.


In this blog, we looked upon φ-RTN, a low-cost grammar-backed language model that interfaces with FST decoders and enhances coverage for long-tail entities in VA ASR settings. φ-RTN eliminates explicit non-terminal expansion at creation time and keeps a single sub-graph per non-terminal, resulting in a small disc footprint. Despite constraints, φ-RTN is complimentary to language models trained on the extended grammar, allowing us to profit from both models and improving WER on long tail entity searches by 10%. Future work will entail expanding φ-RTN to non-English languages with specific grammatical agreements and morphology.

Latest Blogs
This is a decorative image for: Comparison between Cloud-Based and On Premises GPUs
October 6, 2022

Comparison between Cloud-Based and On Premises GPUs

Cloud GPUs vs On Premises GPUs

Cloud GPUs are typically more powerful than on-premises GPU instances. The cost of renting a cloud GPU is generally lower than the cost of purchasing an on-premise GPU. 

Cloud platforms offer fast access to high performance compute and deep learning algorithms, which makes it simpler to start using machine learning models and get early insights into your data. 

Cloud GPUs are better for machine learning because they have lower latency, which is important because the time it takes a neural network to learn from data affects its accuracy. Furthermore, cloud GPUs allow users to take advantage of large-scale training datasets without having to build and maintain their own infrastructure.

On Premises GPUs are better for machine learning if you need high performance or require access to cutting-edge technologies not available in the public cloud. For example, on-premises hardware can be used for deep learning applications that require high memory bandwidth and low latency.

Cloud GPUs: Cloud GPUs are remote data centers where you can rent unused GPU resources. This allows you to run your models on a massive scale, without having to install and manage a local machine learning cluster.

Lower TCO: Cloud GPUs require no upfront investment, making them ideal for companies that are looking to reduce their overall capital expenses. Furthermore, the cost of maintenance and upgrades is also low since it takes place in the cloud rather than on-premises.

Scalability & Flexibility: With cloud-based GPU resources, businesses can scale up or down as needed without any penalty. This ensures that they have the resources they need when demand spikes but also saves them money when there is little or no demand for those resources at all times.

Enhanced Capacity Planning Capabilities: Cloud GPU platforms allow businesses to better plan for future demands by providing estimates of how much processing power will be required in the next 12 months and beyond based on past data points such as workloads run and successes achieved with similar models/algorithms etc... 

Security & Compliance : Since cloud GPUs reside in a remote datacenter separate from your business' core systems, you are ensured peace of mind when it comes to security and compliance matters (eigenvector scanning / firewalls / SELinux etc...) 

Reduced Total Cost Of Ownership (TCO) over time due to pay-as-you-go pricing model which allows you only spend what you actually use vs traditional software licensing models where significant upfront investments are made.

Cloud GPUs: Cloud GPUs offer significant performance benefits over on-premises GPUs. They are accessible from anywhere, and you don't need to own or manage the hardware. This makes them a great choice for data scientists who work with multiple data sets across different platforms.

Numerous Platforms Available for Use: The wide variety of available platforms (Windows, Linux) means that you can run your models using the most popular machine learning libraries and frameworks across different platforms without having to worry about compatibility issues between them.

This is a decorative image for: Impact of the Strong Dollar: Cloud Costs Increasing, Be Indian Buy Indian
October 4, 2022

Impact of the Strong Dollar: Cloud Costs Increasing, Be Indian Buy Indian

Indian SMEs and startups are feeling the effects of the high dollar. These businesses use hyperscalers(MNC Cloud) who cannot modify their rates to account for the changing exchange rate. For certain companies, even a little shift in the currency rate may have a significant effect on their bottom line. Did you know, when the INR-USD exchange rate moved from 60 to 70 in December 2015, it had an impact of around 20% on Digital Innovation?

As the rupee is inching closer to 82 per dollar, the strong dollar has directly impacted the costs of cloud services for Indian businesses. The high cost of storage and computing power, along with bandwidth charges from overseas vendors, has led to a huge increase in the effective rate of these services. This is especially true for startups and SMEs that rely on cloud computing to store and process user data. With the strong dollar continuing to impact the cost of cloud services, it is essential for Indian companies to evaluate their options and adopt local alternatives wherever possible. This blog post will discuss how the strong dollar impacts cloud costs, as well as potential Indian alternatives you can explore in response to this global economic trend. 

What is a Strong Dollar?

A strong US dollar($) is a term used to describe a situation where a US’s currency has appreciated in value compared to other major currencies. This can be due to a variety of factors, including interest rate changes, a country’s current account deficit, and investor sentiment. When a currency appreciates, it means that it is worth more. A strong dollar makes imports more expensive, while making exports cheaper. Strong dollars have been a growing trend in the past couple of years. As the US Federal Reserve continues to hike interest rates, the dollar strengthens further. The rising value of the dollar means that the cost of cloud services, especially from hyperscalers based in the US, will rise as well. 

Increase in Cloud Costs Due to Strong Dollar

Cloud services are essential for modern businesses, as they provide easy access to software, storage, and computing resources. Cloud services are delivered over the internet and are typically charged on a per-use basis. This makes them incredibly convenient for businesses, as they can pay for only the resources they actually use. Cloud computing allows businesses to scale their resources up or down, depending on their current business needs. This makes it suitable for startups, where demand is uncertain, or large enterprises with global operations. Cloud computing is also inherently scalable and allows businesses to quickly react to changing business needs. Cloud computing is a very competitive industry and providers offer attractive prices to attract customers. However, these prices have been impacted by the strong dollar. The dollar has strengthened by 15-20% against the Indian rupee in the last few years. As a result, the costs of services such as storage and bandwidth have increased for Indian companies. Vendors charge their Indian customers in Indian rupees, taking into account the exchange rate. This has resulted in a significant rise in the costs of these services for Indian companies.

Why are Cloud Services Becoming More Expensive?

Cloud services are priced in US dollars. When the dollar is strong, the effective price of services will be higher in Indian rupees, as the cost is not re-adjusted. There are a couple of reasons for this price discrepancy. First, Indian customers will have to pay the same prices as American customers, despite a weaker Indian rupee. Second, vendors have to ensure that they make a profit.

Possible Indian Alternatives to Cloud Services

If you're looking for a cost-effective substitute for services provided by the U.S.-based suppliers, consider E2E Cloud, an Indian cloud service provider. When it comes to cloud services, E2E Cloud provides everything that startups and SMEs could possibly need.

The table below lists some of these services and compares their cost against their US equivalents. 

According to the data in the table above, Indian E2E Cloud Services are much cheaper than their American equivalents. The difference in price between some of these options is substantial. When compared to the prices charged by suppliers in the United States, E2E Cloud's bandwidth costs are surprisingly low. Although not all E2E Cloud services will be noticeably less expensive. Using Indian services, however, has an additional, crucial perk: data sovereignty.


The price of cloud services will rise as the US Dollar appreciates. Indian businesses will need to find ways to counteract the strong dollar's impact on their bottom lines. To do this, one must use E2E Cloud. The availability of E2E Cloud services in INR currency is a bonus on top of the already substantial cost savings. An effective protection against the negative effects of a strong dollar.

This is a decorative image for: Actions CEOs can take to get the value in Cloud Computing
September 28, 2022

Actions CEOs can take to get the value in Cloud Computing

It is not a new thing to say that a major transition is on the way. The transition in which businesses will rely heavily on cloud infrastructure rather than having their own physical IT structure. All of this is due to the cost savings and increased productivity that cloud technology brings to these businesses. Each technological advancement comes with a certain level of risk. Which must be handled carefully in order to ensure the long-term viability of the technology and the benefits it provides.

And CEOs are the primary motivators and decision-makers in any major shift or technological migration in the organization. In the twenty-first century, which is a data-driven century, it is up to the company's leader to decide what and how his/her organization will perform, overcome the risk and succeed in the coming days.

In this blog, we are going to address a few of the actions that CEOs can take to get value in cloud Computing.

  1. A Coordinated Effort

As the saying goes, the more you avoid the risk, the closer it gets. So, if CEOs and their management teams have yet to take an active part or give the necessary attention that their migration journey to the cloud requires, now is the best time to start top-team support for the cloud enablement required to expedite digital strategy, digitalization of the organization, 

The CEO's position is critical because no one else can mediate between the many stakeholders involved, including the CIO, CTO, CFO, chief human-resources officer (CHRO), chief information security officer (CISO), and business-unit leaders.

The move to cloud computing is a collective-action challenge, requiring a coordinated effort throughout an organization's leadership staff. In other words, it's a question of orchestration, and only CEOs can wield the baton. To accelerate the transition to the cloud, CEOs should ask their CIO and CTO what assistance they require to guide the business on the path.

     2. Enhancing business interactions 

To achieve the speed and agility that cloud platforms offer, regular engagement is required between IT managers and their counterparts in business units and functions, particularly those who control products and competence areas. CEOs must encourage company executives to choose qualified decision-makers to serve as product owners for each business capability.

  1. Be Agile

If your organization wants to benefit from the cloud, your IT department, if it isn't already, must become more agile. This entails more than simply transitioning development teams to agile product models. Agile IT also entails bringing agility to your IT infrastructure and operations by transitioning infrastructure and security teams from reactive, "ticket-driven" operations to proactive models in which scrum teams create application programme interfaces (APIs) that service businesses and developers can consume.

  1. Recruiting new employees 

CIOs and CTOs are currently in the lead due to their outstanding efforts in the aftermath of the epidemic. The CEOs must ensure that these executives maintain their momentum while they conduct the cloud transformation. 

Also, Cloud technology necessitates the hire of a highly skilled team of engineers, who are few in number but extremely expensive. As a result, it is envisaged that the CHRO's normal hiring procedures will need to be adjusted in order to attract the proper expertise. Company CEOs may facilitate this by appropriate involvement since this will be critical in deciding the success of the cloud transition.

  1. Model of Business Sustainability 

Funding is a critical component of shifting to the cloud. You will be creating various changes in your sector, from changing the way you now do business to utilizing new infrastructure. As a result, you'll have to spend on infrastructure, tools, and technologies. As CEO, you must develop a business strategy that ensures that every investment provides a satisfactory return on investment for your company. Then, evaluate your investments in order to optimise business development and value.

  1. Taking risks into consideration 

Risk is inherent in all aspects of corporate technology. Companies must be aware of the risks associated with cloud adoption in order to reduce security, resilience, and compliance problems. This includes, among other things, engaging in comprehensive talks about the appropriate procedures for matching risk appetite with technological environment decisions. Getting the business to take the correct risk tone will necessitate special attention from the CEO.

It's easy to allow concerns about security, resilience, and compliance to stall a cloud operation. Instead of allowing risks to derail progress, CEOs should insist on a realistic risk appetite that represents the company plan, while situating cloud computing risks within the context of current on-premises computing risks and demanding choices for risk mitigation in the cloud.


In conclusion, the benefits of cloud computing may be obtained through a high-level approach. A smooth collaboration between the CEO, CIO, and CTO may transform a digital transformation journey into a profitable avenue for the company.

CEOs must consider long-term cloud computing strategy and ensure that the organization is provided with the funding and resources for cloud adoption. The right communication is critical in cloud migration: employees should get these communications from C-suite executives in order to build confidence and guarantee adherence to governance requirements. Simply installing the cloud will not provide value for a company. Higher-level executives (particularly the CEO) must take the lead in the digital transformation path.

Build on the most powerful infrastructure cloud

A vector illustration of a tech city using latest cloud technologies & infrastructure