BUG: Compat with ChatTTS 0.2.1 (#2520)

xorbitsai · Nov 6, 2024 · 8f084b8 · 8f084b8
1 parent 325626f
commit 8f084b8
Show file tree

Hide file tree

Showing 6 changed files with 10 additions and 7 deletions.
diff --git a/.github/workflows/python.yaml b/.github/workflows/python.yaml
@@ -166,7 +166,7 @@ jobs:
             ${{ env.SELF_HOST_PYTHON }} -m pip install -U WeTextProcessing<1.0.4
             ${{ env.SELF_HOST_PYTHON }} -m pip install -U librosa
             ${{ env.SELF_HOST_PYTHON }} -m pip install -U xxhash
-            ${{ env.SELF_HOST_PYTHON }} -m pip install -U "ChatTTS>=0.2,!=0.2.1"
+            ${{ env.SELF_HOST_PYTHON }} -m pip install -U "ChatTTS>=0.2.1"
             ${{ env.SELF_HOST_PYTHON }} -m pip install -U HyperPyYAML
             ${{ env.SELF_HOST_PYTHON }} -m pip uninstall -y matcha-tts
             ${{ env.SELF_HOST_PYTHON }} -m pip install -U onnxruntime-gpu==1.16.0; sys_platform == 'linux'

diff --git a/setup.cfg b/setup.cfg
@@ -111,7 +111,7 @@ all =
     librosa  # For ChatTTS
     xxhash  # For ChatTTS
     torchaudio  # For ChatTTS
-    ChatTTS>=0.2,!=0.2.1
+    ChatTTS>=0.2.1
     lightning>=2.0.0  # For CosyVoice, matcha
     hydra-core>=1.3.2  # For CosyVoice, matcha
     inflect  # For CosyVoice, matcha
@@ -194,7 +194,7 @@ audio =
     librosa
     xxhash
     torchaudio
-    ChatTTS>=0.2,!=0.2.1
+    ChatTTS>=0.2.1
     tiktoken  # For CosyVoice, openai-whisper
     torch>=2.0.0  # For CosyVoice, matcha
     lightning>=2.0.0  # For CosyVoice, matcha

diff --git a/xinference/deploy/docker/requirements.txt b/xinference/deploy/docker/requirements.txt
@@ -50,7 +50,7 @@ nemo_text_processing<1.1.0  # 1.1.0 requires pynini==2.1.6.post1
 WeTextProcessing<1.0.4  # 1.0.4 requires pynini==2.1.6
 librosa  # For ChatTTS
 torchaudio  # For ChatTTS
-ChatTTS>=0.2,!=0.2.1
+ChatTTS>=0.2.1
 xxhash  # For ChatTTS
 torch>=2.0.0  # For CosyVoice
 lightning>=2.0.0  # For CosyVoice, matcha

diff --git a/xinference/deploy/docker/requirements_cpu.txt b/xinference/deploy/docker/requirements_cpu.txt
@@ -47,7 +47,7 @@ nemo_text_processing<1.1.0  # 1.1.0 requires pynini==2.1.6.post1
 WeTextProcessing<1.0.4  # 1.0.4 requires pynini==2.1.6
 librosa  # For ChatTTS
 torchaudio  # For ChatTTS
-ChatTTS>=0.2,!=0.2.1
+ChatTTS>=0.2.1
 xxhash  # For ChatTTS
 torch>=2.0.0  # For CosyVoice
 lightning>=2.0.0  # For CosyVoice, matcha

diff --git a/xinference/model/audio/model_spec.json b/xinference/model/audio/model_spec.json
@@ -127,7 +127,7 @@
     "model_name": "ChatTTS",
     "model_family": "ChatTTS",
     "model_id": "2Noise/ChatTTS",
-    "model_revision": "3b34118f6d25850440b8901cef3e71c6ef8619c8",
+    "model_revision": "1a3c04a8b0651689bd9242fbb55b1f4b5a9aef84",
     "model_ability": "text-to-audio",
     "multilingual": true
   },

diff --git a/xinference/model/audio/tests/test_chattts.py b/xinference/model/audio/tests/test_chattts.py
@@ -25,7 +25,10 @@ def test_chattts(setup):
     client = Client(endpoint)
 
     model_uid = client.launch_model(
-        model_name="ChatTTS", model_type="audio", compile=False
+        model_name="ChatTTS",
+        model_type="audio",
+        compile=False,
+        download_hub="huggingface",
     )
     model = client.get_model(model_uid)
     input_string = (