Homepage | About Us | Careers | Free Trial | Developer Docs | GitHub | GitLab | LinkedIn
Why Choose Speechmatics?
We have a mindset that voice technology should be an accurate representation of the world around us. This has enabled us to build the most accurate and inclusive speech-to-text engine available.
Speechmatics ❤ Open Source
speechmatics-python, a Python client library and CLI for Speechmatics Realtime ASR v2 API
Some of our research 👩‍🔬
Hierarchical Quantized Autoencoders (code published in hqa repo)
NeurIPs 2020 - Will Williams, Sam Ringer, Tom Ash, John Hughes, David MacLeod, Jamie Dougherty. February 19, 2020.
Texture Bias Of CNNs Limits Few-Shot Classification Performance
NeurIPS 2019 - Sam Ringer, Will Williams, Tom Ash, Remi Francis, David MacLeod. October 18, 2019.
Forward-backward retraining of recurrent neural networks
This presents the first “end-to-end” training paper for tasks such as speech recognition. A. Senior and A.J. Robinson. Advances in Neural Information Processing Systems 8, 1996
A recurrent error propagation network speech recognition system
The first application of recurrent nets to speech recognition. A.J. Robinson and F. Fallside. Computer Speech and Language, 5(3):259–274, July 1991.
Some of our tech blogs ✍
How to Build a Streaming DataLoader with PyTorch
How to Boost Emotion Recognition Performance in Speech Using Contrastive Predictive Coding
How to write Kubernetes custom controllers in Go