E2 TTS

Training compute

4.9×10²⁰ FLOP

Parameters

335M

Published

Sep 12, 2024

E2 TTS is an AI model developed by Microsoft (United States), first published in September 2024. It works in the speech domain, on tasks such as text-to-speech (tts) and speech synthesis.

Training it took an estimated 4.9×10²⁰ FLOP of compute (estimation method: operation counting). The model has 335,000,000 parameters. It was trained on roughly 245.8B datapoints.

Access: Unreleased. Its weights are not openly released. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Microsoft
Country of organization: United States
Domain: Speech
Task: Text-to-speech (TTS), Speech synthesis
Training compute: 4.9×10²⁰ FLOP
Compute estimation method: Operation counting
Parameters: 335,000,000
Dataset size: 245.8B
Model accessibility: Unreleased
Open weights: No
Epoch confidence: Confident