Conformer-2: a state-of-the-art speech recognition model trained on 1.1M hours of data
AssemblyAI
JULY 18, 2023
To build generative AI applications leveraging spoken data, product and development teams will need accurate speech to text as a critical component of their AI pipeline. We’re excited to see the innovative AI products built with the improved results from our Conformer-2 model. Research Model Ensembling Conformer-1 utilized a technique called noisy student-teacher training (NST) [3] to achieve state-of-the-art speech recognition.
Let's personalize your content