Update: Empirically Pruned Step Sampling

src/f5_tts/model/cfm.py

@@ -22,6 +22,7 @@
 from f5_tts.model.utils import (
     default,
     exists,
+    get_epss_timesteps,
     lens_to_mask,
     list_str_to_idx,
     list_str_to_tensor,
@@ -92,6 +93,7 @@ def sample(
         seed: int | None = None,
         max_duration=4096,
         vocoder: Callable[[float["b d n"]], float["b nw"]] | None = None,  # noqa: F722
+        use_epss=True,
         no_ref_audio=False,
         duplicate_test=False,
         t_inter=0.1,
@@ -190,7 +192,10 @@ def fn(t, x):
             y0 = (1 - t_start) * y0 + t_start * test_cond
             steps = int(steps * (1 - t_start))
 
-        t = torch.linspace(t_start, 1, steps + 1, device=self.device, dtype=step_cond.dtype)
+        if t_start == 0 and use_epss:  # use Empirically Pruned Step Sampling for low NFE
+            t = get_epss_timesteps(steps, device=self.device, dtype=step_cond.dtype)
+        else:
+            t = torch.linspace(t_start, 1, steps + 1, device=self.device, dtype=step_cond.dtype)
         if sway_sampling_coef is not None:
             t = t + sway_sampling_coef * (torch.cos(torch.pi / 2 * t) - 1 + t)
 

src/f5_tts/model/utils.py

@@ -189,3 +189,22 @@ def repetition_found(text, length=2, tolerance=10):
         if count > tolerance:
             return True
     return False
+
+
+# get the empirically pruned step for sampling
+
+
+def get_epss_timesteps(n, device, dtype):
+    dt = 1 / 32
+    predefined_timesteps = {
+        5: [0, 2, 4, 8, 16, 32],
+        6: [0, 2, 4, 6, 8, 16, 32],
+        7: [0, 2, 4, 6, 8, 16, 24, 32],
+        10: [0, 2, 4, 6, 8, 12, 16, 20, 24, 28, 32],
+        12: [0, 2, 4, 6, 8, 10, 12, 14, 16, 20, 24, 28, 32],
+        16: [0, 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 16, 20, 24, 28, 32],
+    }
+    t = predefined_timesteps.get(n, [])
+    if not t:
+        return torch.linspace(0, 1, n + 1, device=device, dtype=dtype)
+    return dt * torch.tensor(t, device=device, dtype=dtype)