I am a researcher at College of Engineering Trivandrum working in the area of Automatic Speech Recognition under the supervision of Dr. A. R. Jayan and Dr. Rajeev Rajan. I am registered at A. P. J. Abdul Kalam Technological University, Kerala. I love to share what I learn and hence this blog.

I used to be a teacher in the field of electronics, signal processing and digital communication. I was employed as Assistant Professor at Govt. Engineering college, Sreekrishnapuram, Palakkad.

I am involved in various projects on Malayalam language computing. An attempt to analyse Malayalam text phonetically using finite state transducers can be seen here. I have had my hands wet in digital type design and open type engineering of a couple of fonts maintained by Swathanthra Malayalam Computing.

I try to engage with the world immediate to me in the possible ways to make life better for everyone around. It was a moment of pride that upon a request filed by me to the Chief Minister of Kerala, the Director of Higher Education Kerala issued a circular clarifying that there exists no regulations to insist the female teaching faculties to adhere to the dress code of saree.

In this blog I write about myself, my projects and whatever I feels like sharing with the world.

Mozhi Malayalam TTS powered by Mlphon and Mlmorph

Krishna Sankar recently developed a Malayalam - English bilingual Text to Speech System, Mozhi. Check the web demo page and play around with your choice of words and listen to the natural speech it produces. Krishna has set a future goal of understanding the emotional content of the text and read it out accordingly. This is expected to make the application suitable for audio books. Generating audio for arbitrary speaker with very few training samples is another area he plans to work on. [Read More]

Publishing Malayalam Speech Recognition Model

Malayalam speech recognition model trained on various openly available speech and text corpora using Kaldi toolkit is now released here. It is now available for testing on the Vosk-Browser Speech Recognition Demo website. This Malayalam model can be used with Vosk speech recognition toolkit which has bindings for Java, Javascript, C# and Python. A speech recognition architecture that works best in scenarios of limited speech data availability is called a pipeline model, where it is composed of an acoustic model, a language model and a phonetic lexicon. [Read More]

The language of infinite words

I wrote this article for submission to a popular science writing competition by the Department of Science and Technology. Here is my research story, which did not make to grab the award. The story is based on our research on the morphological richness of Malayalam language. How many words are there in your language? How many of them do you know? Can you find all those words in a dictionary? [Read More]

Text Speech and Dialogue: TSD 2020

I presented a paper on Quantitative Analysis of the Morphological Complexity of Malayalam Language at 23rd International Conference on Text, Speech and Dialogue: TSD 2020, Brno, Czech Republic, September 8–11 2020. The year being 2020, the entire conference happened in remote participation mode. Conference proceedings and pre-recorded presentation videos were made available to the participants and we discussed it over online zoom sessions. It was a novel experience and I am super excited about how I got feedbacks and ideas to work on, even after the live sessions. [Read More]

A quantitative analysis of the morphological complexity of Malayalam

My work, Quantitative Analysis of the Morphological Complexity of Malayalam Language is accepted for presentation at the 23rd International Conference on Text, Speech and Dialogue to be held from September 8-10, 2020. The conference proceedings is published by Springer and you can read the paper here. I will share the presentation slides and video after the conference. See the associated data and code here. This blog-post is a bit detailed discussion of the following aspects, analysed in the paper: [Read More]

Releasing Malayalam Speech Corpus

Originally Published in SMC Blog SMC announces the release of Malayalam Speech Corpus (MSC). It is the repository of curated speech samples collected using MSC web application. Speech samples are selected on the criteria that they have at least 3 positive reviews. MSC is a project launched by SMC to crowd source Malayalam speech samples from any contributor who can read out sentences and record them as speech samples. [Read More]

Talks on Speech Recognition Research and Malayalam Computing

Sharing the videos of two informal interviews I did during the past few months. In this video I talk with Hrishikesh Bhaskaran on my involvement with SMC and my projects. This was the part of an interview series hosted by Tinker Hub Foundation. In the following video I talk on Speech recognition systems in general, and on the voice corpus initiative by SMC. This interview is hosted by Mujeeb for IB Computing Youtube channel. [Read More]

A few bits of Information

ഒരു പുസ്തകം മുഴുവനായി വായിച്ചു തീര്‍ക്കുക എന്ന സംഭവം നടന്നിട്ട് ഒരു കൊല്ലത്തിലേറെയായിരുന്നു. ഇന്ന് അത് നടന്നു. മാസങ്ങള്‍ക്കു മുമ്പേ പാതിവഴിയില്‍ നിന്നുപോയ ഒരു പുസ്തകത്തിന് രണ്ടുദിവസം മുമ്പ് വീണ്ടും തുടര്‍ച്ച കണ്ടെടുക്കുകയായിരുന്നു. പോപ്പുലര്‍ സയന്‍സ് ഗണത്തില്‍ പെടുന്ന The Information: A History, a Theory, a Flood എന്ന പുസ്തകമായിരുന്നു അത്. ജെയിംസ് ഗ്ലേക്ക് ആണ് എഴുതിയത്. ഇത് പുസ്തക റിവ്യൂ അല്ല. പക്ഷേ എഞ്ചിനീയറിങ്ങ് ക്ലാസ്സില്‍ പറഞ്ഞ് കേള്‍ക്കാതെ പോകാറുള്ള പലതും ഇതില്‍ കണ്ടു. പ്രത്യേകിച്ചും കണ്ടുപിടുത്തങ്ങളുടെ ചരിത്രവും, സാഹചര്യവും. അദ്ധ്യാപനത്തില്‍ നിന്നും ഒരു ഇടവേളയിലാണ് ഇപ്പോള്‍. ക്ലാസ്സില്‍ സാധാരണ പറയാറില്ലാത്ത ചില ചരിത്രപാഠങ്ങള്‍ ഈ പുസ്തകത്തോട് ചേര്‍ത്ത് ഇവിടെ എഴുതുകയാണ്. [Read More]

Phonetic description of Malayalam consonants

The orthography (system for writing a language) of Malayalam is considered phonemic in nature. It means the graphemes (written symbols) correspond to the phonemes (significant spoken sounds) of the language. But the correspondence between graphemes and phonemes is not precisely one-to-one. The pronunciation of graphemes can depend on its position in a word (word beginning, middle or end) and its proximity to other graphemes. It was two years back, I started to work on a grapheme to phoneme conversion tool for Malayalam. [Read More]

നിർമിതബുദ്ധിയുടെ കാലത്തെ ഭാഷാസാങ്കേതികവിദ്യ

Photo by Markus Spiske on Unsplash ജനയുഗം ഓണപ്പതിപ്പിനു വേണ്ടിയെഴുതി പ്രസിദ്ധീകരിച്ച ലേഖനം ആമുഖം ഭാഷയുടെ ഉപയോഗത്തെ സാങ്കേതികവിദ്യ വളരെയധികം സ്വാധീനിക്കുന്നുണ്ട്. കൈയിലൊതുങ്ങുന്ന കമ്പ്യൂട്ടിങ്ങ് ഉപകരണങ്ങൾ നിത്യജീവിതത്തിന്റെ ഭാഗമാകുന്ന കാലമാണിത്. അപ്പോൾ മനുഷ്യരോടെന്നപോലെ സ്വാഭാവികമായി അവയോടും സംവദിക്കുന്നതിന്റെ ആവശ്യം വർദ്ധിക്കുന്നു. ആപ്പിളിന്റെ സിരിയും, ആമസോണിന്റെ അലക്സയും ഡിജിറ്റൽ അസിസ്റ്റന്റുകളായി സേവനം തുടങ്ങിക്കഴിഞ്ഞു. വെറും വാചാനിർദ്ദേശങ്ങൾ കൊണ്ട് നമുക്കായി സിനിമാടിക്കറ്റ് ബുക്ക് ചെയ്യാനും, ഭക്ഷണം ഓർഡർ ചെയ്യാനും, മെയിലയക്കാനും, അലാറം വെയ്ക്കാനുമൊക്കെ ഇത്തരം ഡിജിറ്റൽ അസിസ്റ്റന്റുകൾക്ക് ഇന്ന് കഴിയും. വളരെ കണിശമായ പ്രോഗ്രാമിങ്ങ് നിർദ്ദേശങ്ങൾ പ്രകാരം മാത്രം പ്രവർത്തിക്കുന്ന ഉപകരണങ്ങളൊക്കെ ഇവയ്ക്കുമുന്നിൽ വഴിമാറുകയാണ്. [Read More]