From 9efcdd2a16cc2d4c193fc3aab65faae1c37216f8 Mon Sep 17 00:00:00 2001 From: Sabil M Akbar Date: Tue, 11 Feb 2025 18:14:00 +0700 Subject: [PATCH] fix:fix cache_kwargs on inference using cache --- parler_tts/modeling_parler_tts.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/parler_tts/modeling_parler_tts.py b/parler_tts/modeling_parler_tts.py index a63c370..31f863f 100644 --- a/parler_tts/modeling_parler_tts.py +++ b/parler_tts/modeling_parler_tts.py @@ -3289,7 +3289,7 @@ def _get_cache(self, cache_implementation: str, max_batch_size: int, max_cache_l cache_dtype = self.dtype cache_kwargs = { "config": self.config.decoder, - "max_batch_size": max_batch_size, + "batch_size": max_batch_size, "max_cache_len": max_cache_len, "device": self.device, "dtype": cache_dtype,