Home

Awesome

Local Audio Server for Yomitan

<sup>Announcement: A new audio collection was released on 2023/06/11! See how to upgrade to the new collection here ("Updating to the new audio collection") </sup>

This Anki add-on runs a local server that Yomitan fetches audio files from, using a database containing over 250,000 unique expressions. With this setup, you are able to create Anki cards nearly instantaneously, get word audio without a working internet connection, and increase the quality and coverage of word audio.

Core maintainer: @Aquafina-water-bottle

<sup> P.S. Feel free to check out <a href="https://aquafina-water-bottle.github.io/jp-mining-note/jpresources/">my other resources</a> if you're interested! </sup>

Reasons for and against this setup

<details> <summary><b>Advantages:</b> <i>(click here)</i> </summary>

  1. Most audio is gotten in almost instantly. Without the local audio server, fetching the audio can take anywhere from one second to a full minute (on particularly bad days).

    Most of the delay from Yomitan when creating cards is from fetching the audio. In other words, audio fetching is the main bottleneck when creating Anki cards. This add-on removes the aforementioned bottleneck, meaning you can make cards with virtually 0 delay.

  2. If you do not have internet access, you can still add audio to your cards.

  3. Compared to standard Yomitan, this improves audio coverage because it adds various sources not covered by Yomitan: Forvo (select users), NHK 2016, and Shinmeikai 8.

  4. Much pre-processing has been done to this audio to make it as high quality as possible:

    • All audio is normalized, so the volume remains relatively similar for each file.
    • Silence has been trimmed from the beginning and end of each file.
    • Using JMdict's data, variant forms with the same readings are back-filled with existing audio.
</details> <details> <summary><b>Disadvantages:</b> <i>(click here)</i> </summary>

  1. This setup requires about 3-5 GB of free space.

  2. This setup requires Anki to be open (or the server to be run manually) in order for Yomitan to fetch audio from it. Note that Yomitan can still fall back to its default sources if the local audio server is unavailable or does not have the requested word.

</details>

Steps

These instructions setup the local audio server for the PC (Windows, MacOS, Linux) release of Anki. If you wish to use this on Android, see here. There is currently no way of using this on AnkiMobile (iOS).

  1. Download all the required audio files. You have two main options:

    1. <details> <summary><b>Ogg/Opus audio (2.5 GiB) (Recommended)</b></summary>

      The Opus audio codec provides much better quality at lower bitrates (which saves a lot of space and makes syncing large collections faster).

      However, Opus is NOT compatible with AnkiMobile (iOS), Android 4, and AnkiWeb. If you use any of these, please use the 2nd option (MP3 audio) below.

      • For Japanese audio, download the files from this torrent. Alternatively, use the magnet link below:

        <details><summary>Magnet link</summary> 
        magnet:?xt=urn:btih:ef90ec428e6abcd560ffc85a2a1c083e0399d003&dn=local-yomichan-audio-collection-2023-06-11-opus.tar.xz&tr=http%3a%2f%2fanidex.moe%3a6969%2fannounce&tr=http%3a%2f%2fnyaa.tracker.wf%3a7777%2fannounce&tr=udp%3a%2f%2fexodus.desync.com%3a6969%2fannounce&tr=udp%3a%2f%2ftracker.opentrackr.org%3a1337%2fannounce&tr=udp%3a%2f%2fopen.stealth.si%3a80%2fannounce&tr=udp%3a%2f%2ftracker.tiny-vps.com%3a6969%2fannounce&tr=udp%3a%2f%2ftracker.moeking.me%3a6969%2fannounce&tr=udp%3a%2f%2fopentracker.i2p.rocks%3a6969%2fannounce&tr=udp%3a%2f%2ftracker.openbittorrent.com%3a6969%2fannounce&tr=udp%3a%2f%2ftracker.torrent.eu.org%3a451%2fannounce&tr=udp%3a%2f%2fexplodie.org%3a6969%2fannounce&tr=udp%3a%2f%2ftracker.zerobytes.xyz%3a1337%2fannounce
        </details>

      • For audio of other languages, use the magnet link below containing Forvo audio and select the zip file of the language you want under the folder Forvo_pronunciation > export > opus > [iso 639 code of language that you want to learn.zip](https://en.wikipedia.org/wiki/List_of_ISO_639_language_codes):

        <details><summary>Magnet link</summary> 
        magnet:?xt=urn:btih:255A1B495719DD3CE0CF8AA708FE6837AD9F57E1&tr=http%3A%2F%2Fbt2.t-ru.org%2Fann%3Fmagnet&dn=%D0%9D%D0%B0%D0%B1%D0%BE%D1%80%20%D0%BF%D1%80%D0%BE%D0%B8%D0%B7%D0%BD%D0%BE%D1%88%D0%B5%D0%BD%D0%B8%D0%B9%20%D0%BE%D1%82%20Forvo.com%20%D0%B4%D0%BB%D1%8F%20GoldenDict%20%D0%BD%D0%B0%20%D0%B2%D1%81%D0%B5%D1%85%20%D1%8F%D0%B7%D1%8B%D0%BA%D0%B0%D1%85
        </details>
      </details>
    2. <details> <summary>MP3 audio (4.9 GiB)</summary>

      Older and less efficient codec, but needed for compatibility with pretty much all devices.

      • For Japanese audio, download the files from this torrent. Alternatively, use the magnet link below:

        <details><summary>Magnet link</summary> 
        magnet:?xt=urn:btih:5bd0aa89667860e68b31a585dc6e7a2bfc811702&dn=local-yomichan-audio-collection-2023-06-11-mp3.tar.xz&tr=http%3a%2f%2fanidex.moe%3a6969%2fannounce&tr=http%3a%2f%2fnyaa.tracker.wf%3a7777%2fannounce&tr=udp%3a%2f%2fexodus.desync.com%3a6969%2fannounce&tr=udp%3a%2f%2ftracker.opentrackr.org%3a1337%2fannounce&tr=udp%3a%2f%2fopen.stealth.si%3a80%2fannounce&tr=udp%3a%2f%2ftracker.tiny-vps.com%3a6969%2fannounce&tr=udp%3a%2f%2ftracker.moeking.me%3a6969%2fannounce&tr=udp%3a%2f%2fopentracker.i2p.rocks%3a6969%2fannounce&tr=udp%3a%2f%2ftracker.openbittorrent.com%3a6969%2fannounce&tr=udp%3a%2f%2ftracker.torrent.eu.org%3a451%2fannounce&tr=udp%3a%2f%2fexplodie.org%3a6969%2fannounce&tr=udp%3a%2f%2ftracker.zerobytes.xyz%3a1337%2fannounce
        </details>

      • For audio of other languages, use the magnet link below containing Forvo audio and select the zip file of the language you want under the folder Forvo_pronunciation > export > opus > [iso 639 code of language that you want to learn.zip](https://en.wikipedia.org/wiki/List_of_ISO_639_language_codes):

        <details><summary>Magnet link</summary> 
        magnet:?xt=urn:btih:255A1B495719DD3CE0CF8AA708FE6837AD9F57E1&tr=http%3A%2F%2Fbt2.t-ru.org%2Fann%3Fmagnet&dn=%D0%9D%D0%B0%D0%B1%D0%BE%D1%80%20%D0%BF%D1%80%D0%BE%D0%B8%D0%B7%D0%BD%D0%BE%D1%88%D0%B5%D0%BD%D0%B8%D0%B9%20%D0%BE%D1%82%20Forvo.com%20%D0%B4%D0%BB%D1%8F%20GoldenDict%20%D0%BD%D0%B0%20%D0%B2%D1%81%D0%B5%D1%85%20%D1%8F%D0%B7%D1%8B%D0%BA%D0%B0%D1%85
        </details>
      </details>

    If you have never downloaded from a torrent before, I highly recommend using the qBittorrent client.

  2. Extract the .tar.xz (Japanese), or .zip file for other languages.

    • Windows users can use 7zip. Note that for the .tar.xz file, 7zip users must extract the resulting tar file as well.
    • Linux and MacOS users can use either the default GUI archive manager or the tar -xf/unzip command.

  3. Download the add-on:

    • Within Anki, navigate to Tools → Add-ons → Get Add-ons...

    • Copy and paste this code: 1045800357

      image

    This add-on only works for Anki versions 2.1.50 and above. If you are using Anki versions 2.1.49 or below, I highly recommend updating. If you can't do this for whatever reason, the legacy instructions and add-on can be found here.

  4. Move the downloaded audio files to the correct place.

    • Within the same Add-ons window, select the add-on (Local Audio Server for Yomichan).

    • Click View files to the right. Your file explorer should now be under Anki2/addons21/1045800357.

      image

    • For Japanese, copy the entire user_files folder (from the second step) into this folder.

      <details> <summary>Expected file structure for Japanese audio <i>(click here)</i></summary> 
        1045800357
  ├── db_utils.py
  ├── server.py
  ├── ...
  └── user_files
      ├── jmdict_forms.json
      ├── forvo_files
      │   ├── akitomo
      │   │   └── 目的.opus
      │   ├── kaoring
      │   │   └── ...
      │   └── ...
      ├── jpod_files
      │   ├── media
      │   │   ├── 000113d2d8419a26e97eacc0b7cfd675.opus
      │   │   ├── 0001d108dd8f99509769192effc1f9e4.opus
      │   │   └── ...
      │   ├── index.json
      │   └── source_meta.json
      ├── nhk16_files
      │   ├── audio
      │   │   ├── 20170616125910.opus
      │   │   └── ...
      │   └── entries.json
      └── shinmeikai8_files
          ├── media
          │   ├── 00001.opus
          │   ├── 00002.opus
          │   └── ...
          └── index.json
      </details>

    • For other languages, change the name of the folder from the second step to forvo_files and copy the entire folder into the user_files folder.

      <details> <summary>Example file structure for other languages' audio <i>(click here)</i></summary> 
        1045800357
  ├── __init__.py
  ├── __pycache__
  │   ├── ...
  ├── ...
  ├── source
  │   ├── __init__.py
  │   ├── __pycache__
  │   │   ├── __init__.cpython-39.opt-2.pyc
  │   │   ├── ...
  │   ├── ...
  ├── user_files
  │   ├── entries.db
  │   ├── entries_version.txt
  │   └── forvo_files
  │       ├── 030Join
  │       ├── ...
  │       ├── Maevew
  │       ├── ...
  └── ...
      </details>

  5. Add the URL in Yomitan.

    • In Yomitan Settings image, go to:

      Audio → Configure audio playback sources.

    • Set the first source to be Custom URL (JSON).

    • Under the first source, set the URL field to:

      http://localhost:5050/?term={term}&reading={reading}

    • If you have other sources, feel free to re-add them under the first source.

    image

  6. Restart Anki. You should see a message saying "Generating local audio database", and should take some time to finish.

  7. Ensure that everything works. To do this, play some audio from Yomitan. You should notice two things:

    • The audio should be played almost immediately after clicking the play button.

    • After playing the audio, you should be able to see the available sources by right-clicking on the play button.

      Here is an example for 読む:

      image

    Play all the sources from the above (読む) to ensure the sound is properly fetched.

Optional Steps: Backfill cards

DillonWall made a fantastic add-on that can backfill cards from any custom URL, including this local audio server.

Optional Steps: Online Forvo Audio Source

To increase audio coverage, I recommend including an extra Forvo audio source. Although the audio is fetched online, this vastly improves coverage compared to the standard Yomitan sources and this local audio server.

Troubleshooting

These are additional instructions and tips if something doesn't work as expected.

Configuring sources

Config File

If you want even more power, sources can be manually configured using a config file. On top of changing the priority of sources and removing sources, you can do the following:

Config Setup

  1. Within the same Add-ons window, select the add-on (Local Audio Server for Yomichan).

  2. Click View files to the right. Your file explorer should now be under Anki2/addons21/1045800357.

  3. Copy default_config.json into user_files, and rename it as config.json.

    <details> <summary>Expected file structure <i>(click here)</i></summary> 
     1045800357
 ├── db_utils.py
 ├── server.py
 ├── default_config.json
 ├── ...
 └── user_files
     ├── config.json <-- Create this file!
     ├── forvo_files
     │   └── ...
     ├── shinmeikai8_files
     │   └── ...
     ├── jpod_files
     │   └── ...
     └── nhk16_files
         └── ...
    </details>

Config Usage Notes

Running without Anki

If you wish to run the server without Anki, do the following:

git clone https://github.com/yomidevs/local-audio-yomichan.git
cd local-audio-yomichan

# You must fill `plugin/user_files` with the audio files, like with step 3 of the main instructions.
# You can run one of the following OS-specific commands:
mklink /d %LOCALAPPDATA%/local-audio-yomichan %APPDATA%/Anki2/addons21/1045800357/user_files # Windows (requires elavated priviledges)
ln -s ~/.local/share/Anki2/addons21/1045800357/user_files ~/local/share/local-audio-yomichan # Linux
ln -s ~/Library/Application\ Support/Anki2/addons21/1045800357/user_files ~/Library/Application\ Support/local-audio-yomichan # MacOS

# After filling in `plugin/user_files` with the audio files, you can now run the server.
# Ensure you have python 3.9 or above.
python3 run_server.py

Install from Source

Credits & Acknowledgements

A lot of people came together, one way or the other, to get this add-on to where it is today. Huge thanks to everyone who made it happen:

License

MIT

Other

<details> <summary>Notes on Forvo Audio Sourcing <i>(click here)</i></summary> </details> <details> <summary>Some Technical Information on the Audio <i>(click here)</i></summary> </details> <details> <summary>Transferring from the deprecated add-on <i>(click here)</i></summary> </details> <details> <summary>Updating to the new audio collection <i>(click here)</i></summary> </details>