server: add OpenAI compatible TTS server (chunked PCM streaming, WAV one-shot)

2026-06-06 12:47:10 +02:00
parent eda8b59092
commit f3cfa5cf47
16 changed files with 40571 additions and 2 deletions
@@ -521,7 +521,7 @@ Sampling:

 Debug:
  --no-fa                 Disable flash attention
-  --clamp-fp16            Clamp hidden states + V to FP16 range
+  --clamp-fp16            Clamp hidden states to FP16 range
  --dump <dir>            Dump intermediate tensors (f32) to <dir>
 ```