4 ms·
Speech samples in the DeepMind publication post: https://deepmind.com/research/publications/End-to-End-Adversarial-Text-to-Speech https://deepmind.com/research/
by seedless-sensat 6y ago
Speech samples in the DeepMind publication post: https://deepmind.com/research/publications/End-to-End-Adversarial-Text-to-Speech https://deepmind.com/research/publications/End-to-End-Advers...
- ryanmjacobs 6y agoI think this is a better link and should have been posted instead. These sound samples are pretty wild! Scary accurate.
- scotth 6y ago"Ablation: No Monotonic Interpolation" is my favorite — wait for it...
- lunixbochs 6y agoI ran into that with tacotron sometimes too
- stubish 6y agoDo I still hear a bit of 'autotune' in there, or am I projecting? I wonder if DeepMind will tackle emoting; they certainly have something good enough for voice overs in a business context.
- ArtWomb 6y agoThanks! You can also experiment with DeepMind's WaveNet using GCP TTS Service (free tier): https://cloud.google.com/text-to-speech https://cloud.google.com/text-to-speech