Initial test with suno/bark-small

.gitignore

@@ -0,0 +1 @@
+__pycache__*

app.py

@@ -0,0 +1,18 @@
+import gradio as gr
+from transformers import AutoProcessor, AutoModel
+import scipy
+
+processor = AutoProcessor.from_pretrained("suno/bark-small")
+model = AutoModel.from_pretrained("suno/bark-small")
+
+def greet(text):
+    inputs = processor(
+        text=[text],
+        return_tensors="pt",
+    )
+    speech_values = model.generate(**inputs, do_sample=True)
+    scipy.io.wavfile.write("tmp.wav", rate=24000, data=speech_values.cpu().numpy().squeeze())
+    return open("tmp.wav", "rb").read()
+
+iface = gr.Interface(fn=greet, inputs="text", outputs="audio")
+iface.launch()