Add VibeVoice-ASR

pyproject.toml

@@ -4,11 +4,11 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "vibevoice"
-version = "0.0.1"
+version = "1.0.0"
 authors = [
-  { name="vibevoice team", email="vibepod@microsoft.com" },
+  { name="vibevoice team", email="VibeVoice@microsoft.com" },
 ]
-description = "A model for speech generation with an AR + diffusion architecture."
+description = "Open-Source Frontier Voice AI."
 readme = "README.md"
 requires-python = ">=3.9"
 classifiers = [
@@ -18,8 +18,7 @@ classifiers = [
 ]
 dependencies = [
     "torch",
-    "accelerate==1.6.0",
-    "transformers==4.51.3", # we develop this project on transformers==4.51.3, later version may not be compatible
+    "accelerate",
     "llvmlite>=0.40.0",
     "numba>=0.57.0",
     "diffusers",
@@ -30,12 +29,21 @@ dependencies = [
     "ml-collections",
     "absl-py",
     "gradio",
-    "av",
-    "aiortc",
-    "uvicorn[standard]",
-    "fastapi"
 ]
 
+[project.optional-dependencies]
+tts = [
+  "transformers==4.51.3", # we develop this project on transformers==4.51.3, later version may not be compatible
+  "av",
+  "aiortc",
+  "uvicorn[standard]",
+  "fastapi"
+]
+
+asr = [
+  "transformers>=4.51.3", # the versions after 4.51.3 are all support
+  "pydub" # for visualization
+]
 
 [project.urls]
 "Homepage" = "https://github.com/microsoft/VibeVoice"