From c329a8855fe397740a9775e42171a0c903a256c7 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?S=C3=B8ren=20Winkel=20Holm?= <sorenmulli>
Date: Wed, 20 Sep 2023 14:54:17 +0200
Subject: [PATCH] fix: Change w2v2 word delimiter token to space

---
 src/coral_models/wav2vec2.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/coral_models/wav2vec2.py b/src/coral_models/wav2vec2.py
index 71bd95cd..79d19233 100644
--- a/src/coral_models/wav2vec2.py
+++ b/src/coral_models/wav2vec2.py
@@ -115,7 +115,7 @@ def load_processor(self) -> Wav2Vec2Processor:
             pad_token="<pad>",
             bos_token="<s>",
             eos_token="</s>",
-            word_delimiter_token="|",
+            word_delimiter_token=" ",
         )
 
         # Set the `model_max_length` attribute of the tokenizer, if it hasn't been set,