Update pyproject.toml

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "f5-tts"
-version = "1.1.5"
+version = "1.1.7"
 description = "F5-TTS: A Fairytaler that Fakes Fluent and Faithful Speech with Flow Matching"
 readme = "README.md"
 license = {text = "MIT License"}
@@ -14,13 +14,13 @@ classifiers = [
     "Programming Language :: Python :: 3",
 ]
 dependencies = [
-    "accelerate>=0.33.0,!=1.7.0",
+    "accelerate>=0.33.0",
     "bitsandbytes>0.37.0; platform_machine != 'arm64' and platform_system != 'Darwin'",
     "cached_path",
     "click",
     "datasets",
     "ema_pytorch>=0.5.2",
-    "gradio>=3.45.2",
+    "gradio<=5.35.0",
     "hydra-core>=1.3.0",
     "jieba",
     "librosa",

src/f5_tts/infer/examples/multi/story.toml

@@ -13,8 +13,8 @@ output_file = "infer_cli_story.wav"
 [voices.town]
 ref_audio = "infer/examples/multi/town.flac"
 ref_text = ""
+speed = 0.8  # will ignore global speed
 
 [voices.country]
 ref_audio = "infer/examples/multi/country.flac"
 ref_text = ""
-

src/f5_tts/infer/examples/multi/story.txt

@@ -1 +1 @@
-A Town Mouse and a Country Mouse were acquaintances, and the Country Mouse one day invited his friend to come and see him at his home in the fields. The Town Mouse came, and they sat down to a dinner of barleycorns and roots, the latter of which had a distinctly earthy flavour. The fare was not much to the taste of the guest, and presently he broke out with [town] “My poor dear friend, you live here no better than the ants. Now, you should just see how I fare! My larder is a regular horn of plenty. You must come and stay with me, and I promise you you shall live on the fat of the land.” [main] So when he returned to town he took the Country Mouse with him, and showed him into a larder containing flour and oatmeal and figs and honey and dates. The Country Mouse had never seen anything like it, and sat down to enjoy the luxuries his friend provided: but before they had well begun, the door of the larder opened and someone came in. The two Mice scampered off and hid themselves in a narrow and exceedingly uncomfortable hole. Presently, when all was quiet, they ventured out again; but someone else came in, and off they scuttled again. This was too much for the visitor. [country] “Goodbye,” [main] said he, [country] “I’m off. You live in the lap of luxury, I can see, but you are surrounded by dangers; whereas at home I can enjoy my simple dinner of roots and corn in peace.”
+A Town Mouse and a Country Mouse were acquaintances, and the Country Mouse one day invited his friend to come and see him at his home in the fields. The Town Mouse came, and they sat down to a dinner of barleycorns and roots, the latter of which had a distinctly earthy flavour. The fare was not much to the taste of the guest, and presently he broke out with [town] "My poor dear friend, you live here no better than the ants! Now, you should just see how I fare! My larder is a regular horn of plenty. You must come and stay with me, and I promise you you shall live on the fat of the land." [main] So when he returned to town he took the Country Mouse with him, and showed him into a larder containing flour and oatmeal and figs and honey and dates. The Country Mouse had never seen anything like it, and sat down to enjoy the luxuries his friend provided: but before they had well begun, the door of the larder opened and someone came in. The two Mice scampered off and hid themselves in a narrow and exceedingly uncomfortable hole. Presently, when all was quiet, they ventured out again; but someone else came in, and off they scuttled again. This was too much for the visitor. [country] "Goodbye," [main] said he, [country] "I'm off. You live in the lap of luxury, I can see, but you are surrounded by dangers; whereas at home I can enjoy my simple dinner of roots and corn in peace."

src/f5_tts/infer/infer_cli.py

@@ -333,6 +333,7 @@ def main():
         text = re.sub(reg2, "", text)
         ref_audio_ = voices[voice]["ref_audio"]
         ref_text_ = voices[voice]["ref_text"]
+        local_speed = voices[voice].get("speed", speed)
         gen_text_ = text.strip()
         print(f"Voice: {voice}")
         audio_segment, final_sample_rate, spectrogram = infer_process(
@@ -347,7 +348,7 @@ def main():
             nfe_step=nfe_step,
             cfg_strength=cfg_strength,
             sway_sampling_coef=sway_sampling_coef,
-            speed=speed,
+            speed=local_speed,
             fix_duration=fix_duration,
             device=device,
         )

src/f5_tts/model/trainer.py

@@ -149,7 +149,7 @@ class Trainer:
         if self.is_main:
             checkpoint = dict(
                 model_state_dict=self.accelerator.unwrap_model(self.model).state_dict(),
-                optimizer_state_dict=self.accelerator.unwrap_model(self.optimizer).state_dict(),
+                optimizer_state_dict=self.optimizer.state_dict(),
                 ema_model_state_dict=self.ema_model.state_dict(),
                 scheduler_state_dict=self.scheduler.state_dict(),
                 update=update,
@@ -242,7 +242,7 @@ class Trainer:
                     del checkpoint["model_state_dict"][key]
 
             self.accelerator.unwrap_model(self.model).load_state_dict(checkpoint["model_state_dict"])
-            self.accelerator.unwrap_model(self.optimizer).load_state_dict(checkpoint["optimizer_state_dict"])
+            self.optimizer.load_state_dict(checkpoint["optimizer_state_dict"])
             if self.scheduler:
                 self.scheduler.load_state_dict(checkpoint["scheduler_state_dict"])
             update = checkpoint["update"]

src/f5_tts/train/finetune_gradio.py

@@ -178,45 +178,12 @@ def get_audio_duration(audio_path):
     return audio.shape[1] / sample_rate
 
 
-def get_rms(
-    y,
-    frame_length=2048,
-    hop_length=512,
-    pad_mode="constant",
-):  # https://github.com/RVC-Boss/GPT-SoVITS/blob/main/tools/slicer2.py
-    padding = (int(frame_length // 2), int(frame_length // 2))
-    y = np.pad(y, padding, mode=pad_mode)
-
-    axis = -1
-    # put our new within-frame axis at the end for now
-    out_strides = y.strides + tuple([y.strides[axis]])
-    # Reduce the shape on the framing axis
-    x_shape_trimmed = list(y.shape)
-    x_shape_trimmed[axis] -= frame_length - 1
-    out_shape = tuple(x_shape_trimmed) + tuple([frame_length])
-    xw = np.lib.stride_tricks.as_strided(y, shape=out_shape, strides=out_strides)
-    if axis < 0:
-        target_axis = axis - 1
-    else:
-        target_axis = axis + 1
-    xw = np.moveaxis(xw, -1, target_axis)
-    # Downsample along the target axis
-    slices = [slice(None)] * xw.ndim
-    slices[axis] = slice(0, None, hop_length)
-    x = xw[tuple(slices)]
-
-    # Calculate power
-    power = np.mean(np.abs(x) ** 2, axis=-2, keepdims=True)
-
-    return np.sqrt(power)
-
-
 class Slicer:  # https://github.com/RVC-Boss/GPT-SoVITS/blob/main/tools/slicer2.py
     def __init__(
         self,
         sr: int,
         threshold: float = -40.0,
-        min_length: int = 2000,
+        min_length: int = 20000,  # 20 seconds
         min_interval: int = 300,
         hop_size: int = 20,
         max_sil_kept: int = 2000,
@@ -247,7 +214,7 @@ class Slicer:  # https://github.com/RVC-Boss/GPT-SoVITS/blob/main/tools/slicer2.
             samples = waveform
         if samples.shape[0] <= self.min_length:
             return [waveform]
-        rms_list = get_rms(y=samples, frame_length=self.win_size, hop_length=self.hop_size).squeeze(0)
+        rms_list = librosa.feature.rms(y=samples, frame_length=self.win_size, hop_length=self.hop_size).squeeze(0)
         sil_tags = []
         silence_start = None
         clip_start = 0
@@ -301,8 +268,7 @@ class Slicer:  # https://github.com/RVC-Boss/GPT-SoVITS/blob/main/tools/slicer2.
             silence_end = min(total_frames, silence_start + self.max_sil_kept)
             pos = rms_list[silence_start : silence_end + 1].argmin() + silence_start
             sil_tags.append((pos, total_frames + 1))
-        # Apply and return slices.
-        ####音频+起始时间+终止时间
+        # Apply and return slices: [chunk, start, end]
         if len(sil_tags) == 0:
             return [[waveform, 0, int(total_frames * self.hop_size)]]
         else: