From 5c21518870064d7393d96a294ff4e908d6a334f2 Mon Sep 17 00:00:00 2001
From: root <zhangyuekai@foxmail.com>
Date: Fri, 21 Feb 2025 02:31:40 +0000
Subject: [PATCH] add requirements.txt

---
 .../TTS/f5-tts/requirements.txt               | 36 +++++++++++++++++++
 egs/wenetspeech4tts/TTS/prepare.sh            |  1 -
 2 files changed, 36 insertions(+), 1 deletion(-)
 create mode 100644 egs/wenetspeech4tts/TTS/f5-tts/requirements.txt

diff --git a/egs/wenetspeech4tts/TTS/f5-tts/requirements.txt b/egs/wenetspeech4tts/TTS/f5-tts/requirements.txt
new file mode 100644
index 000000000..63f1e237c
--- /dev/null
+++ b/egs/wenetspeech4tts/TTS/f5-tts/requirements.txt
@@ -0,0 +1,36 @@
+# F5-TTS
+accelerate>=0.33.0
+bitsandbytes>0.37.0
+cached_path
+click
+datasets
+ema_pytorch>=0.5.2
+gradio>=3.45.2
+hydra-core>=1.3.0
+jieba
+librosa
+matplotlib
+numpy<=1.26.4
+pydub
+pypinyin
+safetensors
+soundfile
+tomli
+torch>=2.0.0
+torchaudio>=2.0.0
+torchdiffeq
+tqdm>=4.65.0
+transformers
+x_transformers>=1.31.14
+
+# icefall
+kaldialign
+lhotse
+tensorboard
+bigvganinference
+sentencepiece
+sherpa-onnx
+k2
+
+# semantic experiment
+s3tokenizer
diff --git a/egs/wenetspeech4tts/TTS/prepare.sh b/egs/wenetspeech4tts/TTS/prepare.sh
index 7b7cd24b7..f1daa0e62 100755
--- a/egs/wenetspeech4tts/TTS/prepare.sh
+++ b/egs/wenetspeech4tts/TTS/prepare.sh
@@ -142,7 +142,6 @@ fi
 
 if [ $stage -le 7 ] && [ $stop_stage -ge 7 ]; then
   log "Stage 7: Extract cosyvoice2 FSQ token (used by ./f5-tts semantic token experiment)"
-  pip install s3tokenizer
   split_name=("valid" "test" "train")
   for split in "${split_name[@]}"; do
       echo "Processing $split"