Research and innovation is embedded in the culture
Whilst many software companies apply technology that has been invented elsewhere, we do things differently. We believe that to truly excel at something, you need world-class expertise at your fingertips.
That’s why Speechmatics has a core team of researchers who are at the cutting-edge of artificial intelligence, neural networks, machine learning and language models embedded at the heart of the company. Teams are encouraged to innovate and iterate, applying their knowledge and expertise quickly and skillfully to our automatic speech recognition technology. Continual development means that our languages are constantly evolving to provide industry-leading accuracy and performance.
With a culture that does not separate research from development, Speechmatics is constantly looking for opportunities to prototype new capabilities, evolve our language offering and rapidly develop new algorithms to remain at the forefront of ASR development.
If you’re interested in finding out more or getting involved in our research, talk to us.
Speechmatics’ paper was submitted and accepted to the most prestigious ML conference – NeurIPs. The paper is about a type of lossy image compression algorithm based on discrete representation learning, leading to a system that can reconstruct images of high-perceptual quality and retain semantically meaningful features despite very high compression rates.
Speechmatics published the paper at NeurIPS 2019 presenting in the meta learning workshop.
In this paper, Speechmatics demonstrates how you could improve recurrent neural network language models by optimizing for downstream speech recognition accuracy directly, rather than the usual generative approach which tries to model the probability of the next word in a sequence.
Speechmatics published the paper at Workshop on Statistical Machine Translation (WMT) 2018 and presented a translation proof of concept.
At Interspeech 2018 in Hyderabad Speechmatics referred to as one of the most accurate providers of ASR after some evaluations, such as one done by Adobe Research. We demonstrated that our continued focus on innovation and to drive new R&D maintains our position in a growing and increasingly challenging field.
This is the first paper that shows that recurrent net language models scale to give very significant gains in speech recognition and it describes the most powerful models to date and some of the special methods needed to train them.
This paper with Google presents a standard large benchmark so that progress in language modelling may be measured. Prior to this paper there was no open, freely available corpus that was large enough to be representative for modern language modelling tasks.
This paper provides an overview of the 2002 state of the art methods to perform speech recognition using neural networks.
Here we show that speech recognition can be used to find information in audio in much the same way that web pages can be found with a search engine.
Here we fundamentally change the main mechanism in speech recognition to make it both faster and more memory efficient (also US patent 5983180).
This presents the first “end-to-end” training paper for tasks such as speech recognition.
Recurrent nets applied to large vocabulary speech recogntiion for the first time.
Recurrent nets are demonstrated to give the best performing system on a well established phoneme recognition task.
The first application of recurrent nets to speech recognition.
This PhD thesis introduces several key concepts of recurrent networks, several different novel architectures, the algorithms needed to train them and applications to speech recognition, coding, and reinforcement learning/game playing.