diff --git a/README.md b/README.md
index 9471adb6123f84bf8394abba7e9a7863bd1268dd..5239f2f3ddba3a304e9d93170c8130eeeb04b4f6 100644
--- a/README.md
+++ b/README.md
@@ -69,6 +69,16 @@ $ deactivate
 Having a CUDA-capable GPU is optional, but can greatly reduce processing time in some situations.
 
 
+To use the custom phonetic mapping for Japanese Romaji, you need to update manually (for now) the g2p DB (within the venv):
+```bash
+$ cp g2p/mappings/langs/rji/* env/lib/python3.11/site-packages/g2p/mappings/langs/rji/
+
+#Then update :
+$ g2p update
+```
+
+
+
 # Use
 
 ## Autokara
@@ -92,6 +102,21 @@ To execute AutoKara on a (pre-extracted) WAV (or OGG, MP3, ...) vocals file, pas
 $ python autokara.py vocals.wav output.ass --vocals
 ```
 
+To use a phonetic transcription optimized for a specific language, use `--lang` (or `-l`) :
+```bash
+$ python autokara.py vocals.wav output.ass --lang jp
+```
+Available languages are :
+```
+jp : Japanese Romaji (default)
+en : English
+```
+
+Full help for all options is available with :
+```bash
+$ python autokara.py -h
+```
+
 ## Useful scripts
 
 To only extract .wav audio from a MKV file :
diff --git a/requirements.txt b/requirements.txt
index 8b063f5e47c50d3149df2cc4787c7f127c576cce..04299146d72785e9cf25e747f7bcea65dd11d05c 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -21,4 +21,5 @@ torchaudio
 tensorboard
 sortedcontainers
 g2p_en
+g2p
 resampy
\ No newline at end of file