NETtalk reimplementation

Training compute

3.6×10¹⁰ FLOP

Parameters

27.5K

Published

Jun 1, 1990

NETtalk reimplementation is an AI model developed by Oregon State University (United States), first published in June 1990. It works in the speech domain, on tasks such as text-to-speech (tts). It counts among the frontier models: the systems trained with the most compute of their moment.

Training it took an estimated 3.6×10¹⁰ FLOP of compute (estimation method: operation counting). The model has 27,480 parameters. It was trained on roughly 7.2K datapoints.

Epoch AI rates the confidence of this record as confident.

Full record

Organization: Oregon State University
Country of organization: United States
Domain: Speech
Task: Text-to-speech (TTS)
Training compute: 3.6×10¹⁰ FLOP
Compute estimation method: Operation counting
Parameters: 27,480
Dataset size: 7.2K
Epoch confidence: Confident