From 5cde4145a896d4a9973ad6dd9deb41428ac230fd Mon Sep 17 00:00:00 2001 From: Aidan Pine Date: Mon, 13 May 2024 18:41:10 +0000 Subject: [PATCH] refactor!: simplify codebase by removing original hifigan code fixes: https://github.com/roedoejet/EveryVoice/issues/425 --- docs/guides/custom.md | 4 +- .../original_hifigan_helper/__init__.py | 235 ------------------ 2 files changed, 2 insertions(+), 237 deletions(-) delete mode 100644 everyvoice/model/vocoder/original_hifigan_helper/__init__.py diff --git a/docs/guides/custom.md b/docs/guides/custom.md index 54be0848..b322581c 100644 --- a/docs/guides/custom.md +++ b/docs/guides/custom.md @@ -100,10 +100,10 @@ everyvoice train text-to-spec config/{{ config_filename('text-to-spec') }} ## Step 8: Synthesize Speech in Your Language! -You can synthesize by pointing the CLI to your trained feature prediction network and passing in the text. You can export to wav, npy, or pt files. +You can synthesize by pointing the CLI to your trained feature prediction network and passing in the text. You can export the wav or spectrogram (pt) files. ```bash -everyvoice synthesize from-text logs_and_checkpoints/FeaturePredictionExperiment/base/checkpoints/last.ckpt -t "මෙදා සැරේ සාකච්ඡාවක් විදියට නෙවෙයි නේද පල කරල තියෙන්නෙ" -a gpu -d 1 +everyvoice synthesize from-text logs_and_checkpoints/FeaturePredictionExperiment/base/checkpoints/last.ckpt -t "මෙදා සැරේ සාකච්ඡාවක් විදියට නෙවෙයි නේද පල කරල තියෙන්නෙ" -a gpu -d 1 --output-type wav ```