mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-08-09 01:52:41 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix typos

Browse Source
This commit is contained in:
Fangjun Kuang 2024-12-30 18:20:49 +08:00
parent 5181a47775
commit 9fd87c5b3e
7 changed files with 13 additions and 8 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
.github/scripts/baker_zh/TTS/run-matcha.sh vendored
View File

@ -92,7 +92,7 @@ function export_onnx() {
ls -lh *.onnx ls -lh *.onnx
if false; then if false; then
# THe CI machine does not have enough memory to run it # The CI machine does not have enough memory to run it
# #
curl -SL -O https://github.com/csukuangfj/models/raw/refs/heads/master/hifigan/generator_v1 curl -SL -O https://github.com/csukuangfj/models/raw/refs/heads/master/hifigan/generator_v1
curl -SL -O https://github.com/csukuangfj/models/raw/refs/heads/master/hifigan/generator_v2 curl -SL -O https://github.com/csukuangfj/models/raw/refs/heads/master/hifigan/generator_v2

2
.github/scripts/ljspeech/TTS/run-matcha.sh vendored
View File

@ -90,7 +90,7 @@ function export_onnx() {
ls -lh *.onnx ls -lh *.onnx
if false; then if false; then
# THe CI machine does not have enough memory to run it # The CI machine does not have enough memory to run it
# #
curl -SL -O https://github.com/csukuangfj/models/raw/refs/heads/master/hifigan/generator_v1 curl -SL -O https://github.com/csukuangfj/models/raw/refs/heads/master/hifigan/generator_v1
curl -SL -O https://github.com/csukuangfj/models/raw/refs/heads/master/hifigan/generator_v2 curl -SL -O https://github.com/csukuangfj/models/raw/refs/heads/master/hifigan/generator_v2

2
.github/workflows/baker_zh.yml vendored
View File

@ -12,7 +12,7 @@ on:
workflow_dispatch: workflow_dispatch:
concurrency: concurrency:
group: ljspeech-${{ github.ref }} group: baker-zh-${{ github.ref }}
cancel-in-progress: true cancel-in-progress: true
jobs: jobs:

1
egs/baker_zh/TTS/.gitignore vendored
View File

@ -1,4 +1,5 @@
path.sh path.sh
*.onnx
*.wav *.wav
generator_v1 generator_v1
generator_v2 generator_v2

10
egs/baker_zh/TTS/README.md
View File

@ -3,7 +3,11 @@
It is for the dataset from It is for the dataset from
https://en.data-baker.com/datasets/freeDatasets/ https://en.data-baker.com/datasets/freeDatasets/
The dataset contains 10000 Chinese sentences of a native Chinese female speaker. The dataset contains 10000 Chinese sentences of a native Chinese female speaker,
which is about 12 hours.
**Note**: The dataset is for non-commercial use only.
# matcha # matcha
@ -28,7 +32,7 @@ python3 ./matcha/train.py \
To inference, use: To inference, use:
```bash ```bash
# Download Hifigan vocoder. We use Hifigan v1 below. You can select from v1, v2, or v3 # Download Hifigan vocoder. We use Hifigan v2 below. You can select from v1, v2, or v3
wget https://github.com/csukuangfj/models/raw/refs/heads/master/hifigan/generator_v2 wget https://github.com/csukuangfj/models/raw/refs/heads/master/hifigan/generator_v2
@ -70,7 +74,7 @@ python3 ./matcha/export_onnx.py \
--cmvn ./data/fbank/cmvn.json --cmvn ./data/fbank/cmvn.json
``` ```
The above command generate the following files: The above command generates the following files:
``` ```
-rw-r--r-- 1 kuangfangjun root 72M Dec 27 18:53 model-steps-2.onnx -rw-r--r-- 1 kuangfangjun root 72M Dec 27 18:53 model-steps-2.onnx
-rw-r--r-- 1 kuangfangjun root 73M Dec 27 18:54 model-steps-3.onnx -rw-r--r-- 1 kuangfangjun root 73M Dec 27 18:54 model-steps-3.onnx

2
egs/baker_zh/TTS/matcha/export_onnx.py
View File

@ -191,7 +191,7 @@ def main():
"model_author": "icefall", "model_author": "icefall",
"maintainer": "k2-fsa", "maintainer": "k2-fsa",
"dataset": "baker-zh", "dataset": "baker-zh",
"use_eos_bos": 1, "use_eos_bos": 0,
"dataset_url": "https://www.data-baker.com/open_source.html", "dataset_url": "https://www.data-baker.com/open_source.html",
"dataset_comment": "The dataset is for non-commercial use only.", "dataset_comment": "The dataset is for non-commercial use only.",
"num_ode_steps": num_steps, "num_ode_steps": num_steps,

2
egs/ljspeech/TTS/README.md
View File

@ -166,7 +166,7 @@ To export the checkpoint to onnx:
--tokens ./data/tokens.txt --tokens ./data/tokens.txt
``` ```
The above command generate the following files: The above command generates the following files:
- model-steps-2.onnx - model-steps-2.onnx
- model-steps-3.onnx - model-steps-3.onnx