Researcher profile

Anupam Singh

Anupam Singh contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
12works
0followers
12topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

12 published item(s)

preprint2022arXiv

A Simple Baseline for Domain Adaptation in End to End ASR Systems Using Synthetic Data

Automatic Speech Recognition(ASR) has been dominated by deep learning-based end-to-end speech recognition models. These approaches require large amounts of labeled data in the form of audio-text pairs. Moreover, these models are more susceptible to domain shift as compared to traditional models. It is common practice to train generic ASR models and then adapt them to target domains using comparatively smaller data sets. We consider a more extreme case of domain adaptation where text-only corpus is available. In this work, we propose a simple baseline technique for domain adaptation in end-to-end speech recognition models. We convert the text-only corpus to audio data using single speaker Text to Speech (TTS) engine. The parallel data in the target domain is then used to fine-tune the final dense layer of generic ASR models. We show that single speaker synthetic TTS data coupled with final dense layer only fine-tuning provides reasonable improvements in word error rates. We use text data from address and e-commerce search domains to show the effectiveness of our low-cost baseline approach on CTC and attention-based models.

preprint2022arXiv

Powers in finite orthogonal and symplectic groups: A generating function approach

For an integer $M\geq 2$ and a finite group $G$, an element $α\in G$ is called an $M$-th power if it satisfies $A^M=α$ for some $A\in G$. In this article, we will deal with the case when $G$ is finite symplectic or orthogonal group over a field of order $q$. We introduce the notion of $M^*$-power SRIM polynomials. This, amalgamated with the concept of $M$-power polynomial, we provide the complete classification of the conjugacy classes of regular semisimple, semisimple, cyclic and regular elements in $G$, which are $M$-th powers, when $(M,q)=1$. The approach here is of generating functions, as worked on by Jason Fulman, Peter M. Neumann, and Cheryl Praeger in the memoir "A generating function approach to the enumeration of matrices in classical groups over finite fields". As a byproduct, we obtain the corresponding probabilities, in terms of generating functions.

preprint2020arXiv

Algorithms in Linear Algebraic Groups

This paper presents some algorithms in linear algebraic groups. These algorithms solve the word problem and compute the spinor norm for orthogonal groups. This gives us an algorithmic definition of the spinor norm. We compute the double coset decomposition with respect to a Siegel maximal parabolic subgroup, which is important in computing infinite-dimensional representations for some algebraic groups.

preprint2020arXiv

Branching rules for Unitary and Symplectic matrices

This paper concerns the enumeration of simultaneous conjugacy classes of tuples of commuting unitary matrices and of commuting symplectic matrices over a finite field $\mathbf{F}_q$ of odd size. For any given conjugacy class, the orbits for the action of its centralizer group on itself by conjugation are called branches. We determine the branching rules for the unitary groups $U_2(\mathbf{F}_q), U_3(\mathbf{F}_q)$, and for the symplectic groups $Sp_2(\mathbf{F}_q), Sp_4(\mathbf{F}_q)$.

preprint2020arXiv

Dark Energy Gravitational Wave Observations and Ice Age Periodicity

Dark Energy is the dominant component of the energy density of the Universe. However, it is also very elusive since its interaction with the rest of the Universe is primarily gravitational.Since Dark Energy is a low energy phenomenon from the perspective of particle physics and field theory, a fundamental approach based on fields in curved space is sufficient to understand the current dynamics of Dark Energy. The key issue is to understand the gravitational dynamics of Dark Energy and its observational consequences. However, finding the observational consequences of Dark Energy dynamics has been a very challenging task. For something which is the dominant component of the energy density of the Universe, Dark Energy appears to be very distant and reclusive. Here we show that the Dark Energy dynamics results in the production of gravitational waves which produce the ellipticity variation in the orbit of earth that results in the periodicity of the Ice Ages observed and documented by geologists and climatologists. Previously, no observational signature of gravitational waves produced by Dark Energy dynamics has been reported. Further, no interpretation of the ellipticity variation of the orbit of earth due to gravitational waves or the linking of such gravitational waves to the Ice Age periodicity has been reported previously. We hope that the current work will lead to some fresh insights and some more interesting work.

preprint2020arXiv

Finiteness of $z$-classes in reductive groups

Let $k$ be a perfect field such that for every $n$ there are only finitely many field extensions, up to isomorphism, of $k$ of degree $n$. If $G$ is a reductive algebraic group defined over $k$, whose characteristic is very good for $G$, then we prove that $G(k)$ has only finitely many $z$-classes. For each perfect field $k$ which does not have the above finiteness property we show that there exist groups $G$ over $k$ such that $G(k)$ has infinitely many $z$-classes.

preprint2020arXiv

Gravitational Wave Formation from the Collapse of Dark Energy Field Configurations

Dark Energy is the dominant component of the energy density of the universe. In a previous paper, we have shown that the collapse of dark energy fields leads to the formation of Super Massive Black Holes with masses comparable to the masses of Black Holes at the centers of galaxies. Thus it becomes a pressing issue to investigate the other physical consequences of the collapse of Dark Energy fields. Given that the primary interactions of Dark Energy fields with the rest of the Universe are gravitational, it is particularly interesting to investigate the gravitational wave signals emitted during the process of the collapse of Dark Energy fields. This is the focus of the current work described in this paper. We describe and use the 3+1 BSSN formalism to follow the evolution of the dark energy fields coupled with gravity and to extract the gravitational wave signals. Finally, we describe the results of our numerical computations and the gravitational wave signals produced as a result of the collapse of the dark energy fields.

preprint2019arXiv

Computing $n^{\rm th}$ roots in $SL_2$ and Fibonacci polynomials

Let $k$ be a field of characteristic $\neq 2$. In this paper we study squares, cubes and their products in split and anisotropic groups of type $A_1$. In split case, we show that computing $n^{\rm th}$ roots is equivalent to finding solutions of certain polynomial equations in at most two variables over the base field $k$. The description of these polynomials involves generalised Fibonacci polynomials. Using this we obtain asymptotic proportions of $n^{\rm th}$ powers, and conjugacy classes which are $n^{\rm th}$ powers, in ${\rm SL}_2(\mathbb F_q)$ when $n$ is a prime or $n = 4$. We also extend already known Waring type result for ${\rm SL}_2(\mathbb F_q)$, that every element of ${\rm SL}_2(\mathbb F_q)$ is a product of two squares, to ${\rm SL}_2(k)$ for an arbitrary $k$. For anisotropic groups of type $A_1$, namely ${\rm SL}_1(Q)$ where $Q$ is a quaternion division algebra, we prove that when $2$ is a square in $k$, every element of ${\rm SL}_1(Q)$ is a product of two squares if and only if $-1$ is a square in ${\rm SL}_1(Q)$.

preprint2010arXiv

Gravitational Collapse of Dark Energy Field Configurations

It is now a well established fact based on several observations that we live in a Universe dominated by dark energy. The most natural candidates for dark energy are fields in curved space-time. We develop the formalism to study the gravitational collapse of fields given any general potential. We apply this formalism to models of dark energy motivated by particle physics considerations. We solve the resulting evolution equations which determine the time evolution of field configurations as well as the dynamics of spacetime.