## Env for kaldi

git clone https://github.com/kaldi-asr/kaldi.git kaldi --origin upstream

`kaldi/tools/INSTALL`

To check the prerequisites for Kaldi, first run

  extras/check_dependencies.sh

and see if there are any system-level installations you need to do. Check the
output carefully. There are some things that will make your life a lot easier
if you fix them at this stage. If your system default C++ compiler is not
supported, you can do the check with another compiler by setting the CXX
environment variable, e.g.

  CXX=g++-4.8 extras/check_dependencies.sh

Then run

  make

which by default will install ATLAS headers, OpenFst, SCTK and sph2pipe.
OpenFst requires a relatively recent C++ compiler with C++11 support, e.g.
g++ >= 4.7, Apple clang >= 5.0 or LLVM clang >= 3.3. If your system default
compiler does not have adequate support for C++11, you can specify a C++11
compliant compiler as a command argument, e.g.

  make CXX=g++-4.8

If you have multiple CPUs and want to speed things up, you can do a parallel
build by supplying the "-j" option to make, e.g. to use 4 CPUs

  make -j 4

In extras/, there are also various scripts to install extra bits and pieces that
are used by individual example scripts.  If an example script needs you to run
one of those scripts, it will tell you what to do.

`kaldi/src/INSTALL`

These instructions are valid for UNIX-like systems (these steps have
been run on various Linux distributions; Darwin; Cygwin).  For native Windows
compilation, see ../windows/INSTALL.

You must first have completed the installation steps in ../tools/INSTALL
(compiling OpenFst; getting ATLAS and CLAPACK headers).

The installation instructions are

  ./configure --shared
  make depend -j 8
  make -j 8

Note that we added the "-j 8" to run in parallel because "make" takes a long
time.  8 jobs might be too many for a laptop or small desktop machine with not
many cores.

For more information, see documentation at http://kaldi-asr.org/doc/
and click on "The build process (how Kaldi is compiled)".

## Pre-processing

:::info
注意先用sox轉音檔格式，轉成 16 kHz sampling, signed-integer, 16 bits
:::

如果您想将新的文件保存到名为`new_train`和`new_test`的文件夹中，可以使用以下命令：

mkdir new_train
mkdir new_test

for file in train/*.wav; do
  new_file="new_train/$(basename "$file")"
  sox "$file" -r 16000 -e signed-integer -b 16 "$new_file"
done

for file in test/*.wav; do
  new_file="new_test/$(basename "$file")"
  sox "$file" -r 16000 -e signed-integer -b 16 "$new_file"
done

这些命令首先创建了`new_train`和`new_test`文件夹，然后将转换后的音频文件保存到这两个新文件夹中。文件结构和原始文件夹结构相同，但格式已更改。

## Build some files


## `csv_to_txt.py`

import csv

with open('text.txt', 'w') as outfile:
    with open('train-toneless.csv', 'r', errors='ignore') as infile:
        [outfile.write(" ".join(row) + "\n") for row in csv.reader(infile)]
    outfile.close()

## `txt_to_csv.py`

import csv

# Read the text file
with open('your_text_file.txt', 'r') as file:
    lines = file.readlines()

# Parse the text data and convert it into a list of dictionaries
data = []
for line in lines:
    parts = line.strip().split()
    id, *words = parts
    data.append({'id': id, 'text': ' '.join(words)})

# Sort the data by the 'id' field
sorted_data = sorted(data, key=lambda x: int(x['id']))

# Write the sorted data to a CSV file
with open('output.csv', 'w', newline='') as csvfile:
    fieldnames = ['id', 'text']
    writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
    
    writer.writeheader()
    for row in sorted_data:
        writer.writerow(row)

print("Data has been transferred to output.csv and sorted by ID.")

### GPU version

記得 GPU job 都是 1

---

## Build ESPnet

[助教筆記](https://aged-foe-6b2.notion.site/ESPnet-egs2-b40c366ac70646628c4ecf54633780ac)

### 跑yesno

`espnet/eg2/yesno/asr1` 跑 `bash run.sh`

### `data`

**`spk2utt`**

S0002 BAC009S0002W0122 BAC009S0002W0123 BAC009S0002W0124 BAC009S0002W0125 BAC009S0002W0126 BAC009S0002W0127 BAC009S0002W0128 BAC009S0002W0129 

**`text`**

**`utt2spk`**

BAC009S0002W0122 S0002

**`wav.scp`**

BAC009S0002W0122 /home/mllab/Desktop/espnet/egs2/aishell/asr1/downloads/data_aishell/wav/train/S0002/BAC009S0002W0122.wav

---

## HW5

AttributeError: 'HParams' object has no attribute 'duration_discriminator_type'

[Github Issue](https://github.com/p0p4k/vits2_pytorch/issues/64)

Add `"duration_discriminator_type" :"dur_disc_2",` in the `model` from `configs/vits2_ljs_nosdp.json`

`configs/vits2_ljs_nosdp.json` 可以調低 batch_size，原先 64