mirror of https://github.com/coqui-ai/TTS.git
11 lines
458 B
Plaintext
11 lines
458 B
Plaintext
- location sensitive attention
|
|
- attention masking
|
|
x updated audio pre-processing (removed preemphasis and updated normalization)
|
|
- testing with example senteces
|
|
- optional eval-iteration
|
|
- plot alignments and specs for test sentences
|
|
- use lws as an alternative audio processor
|
|
- make configurable audio module
|
|
- make precomputed of on-the-fly dataloader configurable
|
|
- use of preemphasis is configurable. It is skipped if config.preamphasis == 0
|