- Newest
- Most votes
- Most comments
Amazon Nova Sonic has a maximum token context window of 300,000 tokens. This means the model can process and respond to inputs containing up to 300,000 tokens at once.
To put this in perspective, in English, tokens are typically around 4 characters long, with a token being approximately 3/4 of a word. So 300,000 tokens would roughly equate to about 225,000 words, which is a substantial amount of text for processing conversations and interactions.
Amazon Nova Sonic is designed for real-time, human-like voice conversations with features like fluid dialogue, turn-taking, function calling, and knowledge grounding with enterprise data. It currently supports English (including American and British accents), with additional languages planned for the future.
Sources
Amazon Nova - Generative Foundation Model - AWS
Community | How To Choose Your LLM
answered a year ago
Hi,
I couldn't find documentation with specific info on the length of audio token. the closes I have found is a pricing calculation equating the audio time to an average of tokens.
seems that on average once second equates to 25 tokens
answered 10 months ago
Relevant content
asked a year ago
asked a year ago
- AWS OFFICIALUpdated 4 months ago
