New Releases - Five New Languages and Three New Use Case Speech Models
Deepgram is excited to continue executing the vision of having everyone heard and understood through speech recognition. To that end, we are announcing five new languages and three new use case speech models. In addition, we have improved our Hindi and Spanish language models and our Conversational AI and Meetings models. These improvements mean better accuracy for your transcription applications. Through transfer learning backed by our End-to-End Deep Learning architecture and the process of creating conversational audio datasets for training, we have been able to achieve very high accuracy on our new speech models. Our new models are trained on real conversations, not Libraspeech type datasets of people reading audiobooks because no one talks like they are reading audiobooks. This allows our speech models to better transcribe jargon, accents, dialects, code-switching (switching from Hindi to English words), terminology (ROI, EBITDA), and alphanumerics. This means our models provide better real-world transcription accuracy than standard language or general one-size-fits-all speech models. Improved Production Language Models
New Language Models
Our French and German language models are FREE for the next 30 days of use to collect more audio for training and improvement. Sign up here to get early access. New and Improved Use Case Models
Conversational AI (improved)
Because of our ability to tailor these models for your use case, dialect, noise, accents, and terminology within weeks, consider these speech models as a foundation to even better accuracy for your specific use case. Our internal testing with Google and Amazon's speech-to-text solutions show these new and improved models are better out of the gate. Test us out with our self-service Console or let us perform a guided test and train a tailored model for your specific use case.
If you have any feedback about this post, or anything else around Deepgram, we'd love to hear from you. Please let us know in our GitHub discussions .