Merge pull request #217 from idiap/stdout

fix(bin): log to stdout in cli tools

TTS/bin/compute_attention_masks.py

@@ -2,6 +2,7 @@
 import importlib
 import logging
 import os
+import sys
 from argparse import RawTextHelpFormatter
 
 import numpy as np
@@ -18,7 +19,7 @@
 from TTS.utils.generic_utils import ConsoleFormatter, setup_logger
 
 if __name__ == "__main__":
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
 
     # pylint: disable=bad-option-value
     parser = argparse.ArgumentParser(

TTS/bin/compute_embeddings.py

@@ -1,6 +1,7 @@
 import argparse
 import logging
 import os
+import sys
 from argparse import RawTextHelpFormatter
 
 import torch
@@ -102,7 +103,7 @@ def compute_embeddings(
 
 
 if __name__ == "__main__":
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
 
     parser = argparse.ArgumentParser(
         description="""Compute embedding vectors for each audio file in a dataset and store them keyed by `{dataset_name}#{file_path}` in a .pth file\n\n"""

TTS/bin/compute_statistics.py

@@ -5,6 +5,7 @@
 import glob
 import logging
 import os
+import sys
 
 import numpy as np
 from tqdm import tqdm
@@ -18,7 +19,7 @@
 
 def main():
     """Run preprocessing process."""
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stderr, formatter=ConsoleFormatter())
 
     parser = argparse.ArgumentParser(description="Compute mean and variance of spectrogtram features.")
     parser.add_argument("config_path", type=str, help="TTS config file path to define audio processin parameters.")

TTS/bin/eval_encoder.py

@@ -1,5 +1,6 @@
 import argparse
 import logging
+import sys
 from argparse import RawTextHelpFormatter
 
 import torch
@@ -53,7 +54,7 @@ def compute_encoder_accuracy(dataset_items, encoder_manager):
 
 
 if __name__ == "__main__":
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
 
     parser = argparse.ArgumentParser(
         description="""Compute the accuracy of the encoder.\n\n"""

TTS/bin/extract_tts_spectrograms.py

@@ -4,6 +4,7 @@
 import argparse
 import logging
 import os
+import sys
 
 import numpy as np
 import torch
@@ -273,7 +274,7 @@ def main(args):  # pylint: disable=redefined-outer-name
 
 
 if __name__ == "__main__":
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
 
     parser = argparse.ArgumentParser()
     parser.add_argument("--config_path", type=str, help="Path to config file for training.", required=True)

TTS/bin/find_unique_chars.py

@@ -2,6 +2,7 @@
 
 import argparse
 import logging
+import sys
 from argparse import RawTextHelpFormatter
 
 from TTS.config import load_config
@@ -10,7 +11,7 @@
 
 
 def main():
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
 
     # pylint: disable=bad-option-value
     parser = argparse.ArgumentParser(

TTS/bin/find_unique_phonemes.py

@@ -3,6 +3,7 @@
 import argparse
 import logging
 import multiprocessing
+import sys
 from argparse import RawTextHelpFormatter
 
 from tqdm.contrib.concurrent import process_map
@@ -20,7 +21,7 @@ def compute_phonemes(item):
 
 
 def main():
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
 
     # pylint: disable=W0601
     global c, phonemizer

TTS/bin/remove_silence_using_vad.py

@@ -4,6 +4,7 @@
 import multiprocessing
 import os
 import pathlib
+import sys
 
 import torch
 from tqdm import tqdm
@@ -77,7 +78,7 @@ def preprocess_audios():
 
 
 if __name__ == "__main__":
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
 
     parser = argparse.ArgumentParser(
         description="python TTS/bin/remove_silence_using_vad.py -i=VCTK-Corpus/ -o=VCTK-Corpus-removed-silence/ -g=wav48_silence_trimmed/*/*_mic1.flac --trim_just_beginning_and_end"

TTS/bin/synthesize.py

@@ -311,8 +311,9 @@ def parse_args() -> argparse.Namespace:
 
 def main() -> None:
     """Entry point for `tts` command line interface."""
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
     args = parse_args()
+    stream = sys.stderr if args.pipe_out else sys.stdout
+    setup_logger("TTS", level=logging.INFO, stream=stream, formatter=ConsoleFormatter())
 
     pipe_out = sys.stdout if args.pipe_out else None
 

TTS/bin/train_encoder.py

@@ -322,7 +322,7 @@ def main(args):  # pylint: disable=redefined-outer-name
 
 
 if __name__ == "__main__":
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
 
     args, c, OUT_PATH, AUDIO_PATH, c_logger, dashboard_logger = init_training()
 

TTS/bin/train_tts.py

@@ -1,5 +1,6 @@
 import logging
 import os
+import sys
 from dataclasses import dataclass, field
 
 from trainer import Trainer, TrainerArgs
@@ -17,7 +18,7 @@ class TrainTTSArgs(TrainerArgs):
 
 def main():
     """Run `tts` model training directly by a `config.json` file."""
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
 
     # init trainer args
     train_args = TrainTTSArgs()

TTS/bin/train_vocoder.py

@@ -1,5 +1,6 @@
 import logging
 import os
+import sys
 from dataclasses import dataclass, field
 
 from trainer import Trainer, TrainerArgs
@@ -18,7 +19,7 @@ class TrainVocoderArgs(TrainerArgs):
 
 def main():
     """Run `tts` model training directly by a `config.json` file."""
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
 
     # init trainer args
     train_args = TrainVocoderArgs()

TTS/bin/tune_wavegrad.py

@@ -2,6 +2,7 @@
 
 import argparse
 import logging
+import sys
 from itertools import product as cartesian_product
 
 import numpy as np
@@ -17,7 +18,7 @@
 from TTS.vocoder.models import setup_model
 
 if __name__ == "__main__":
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
 
     parser = argparse.ArgumentParser()
     parser.add_argument("--model_path", type=str, help="Path to model checkpoint.")

TTS/encoder/utils/prepare_voxceleb.py

@@ -216,7 +216,7 @@ def processor(directory, subset, force_process):
 
 
 if __name__ == "__main__":
-    setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+    setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
     if len(sys.argv) != 4:
         print("Usage: python prepare_data.py save_directory user password")
         sys.exit()

TTS/server/server.py

@@ -25,7 +25,7 @@
 from TTS.utils.synthesizer import Synthesizer
 
 logger = logging.getLogger(__name__)
-setup_logger("TTS", level=logging.INFO, screen=True, formatter=ConsoleFormatter())
+setup_logger("TTS", level=logging.INFO, stream=sys.stdout, formatter=ConsoleFormatter())
 
 
 def create_argparser() -> argparse.ArgumentParser:

TTS/utils/generic_utils.py

@@ -2,9 +2,10 @@
 import datetime
 import importlib
 import logging
+import os
 import re
 from pathlib import Path
-from typing import Any, Callable, Dict, Optional, TypeVar, Union
+from typing import Any, Callable, Dict, Optional, TextIO, TypeVar, Union
 
 import torch
 from packaging.version import Version
@@ -107,25 +108,34 @@ def setup_logger(
     level: int = logging.INFO,
     *,
     formatter: Optional[logging.Formatter] = None,
-    screen: bool = False,
-    tofile: bool = False,
-    log_dir: str = "logs",
+    stream: Optional[TextIO] = None,
+    log_dir: Optional[Union[str, os.PathLike[Any]]] = None,
     log_name: str = "log",
 ) -> None:
+    """Set up a logger.
+
+    Args:
+        logger_name: Name of the logger to set up
+        level: Logging level
+        formatter: Formatter for the logger
+        stream: Add a StreamHandler for the given stream, e.g. sys.stderr or sys.stdout
+        log_dir: Folder to write the log file (no file created if None)
+        log_name: Prefix of the log file name
+    """
     lg = logging.getLogger(logger_name)
     if formatter is None:
         formatter = logging.Formatter(
             "%(asctime)s.%(msecs)03d - %(levelname)-8s - %(name)s: %(message)s", datefmt="%y-%m-%d %H:%M:%S"
         )
     lg.setLevel(level)
-    if tofile:
+    if log_dir is not None:
         Path(log_dir).mkdir(exist_ok=True, parents=True)
         log_file = Path(log_dir) / f"{log_name}_{get_timestamp()}.log"
         fh = logging.FileHandler(log_file, mode="w")
         fh.setFormatter(formatter)
         lg.addHandler(fh)
-    if screen:
-        sh = logging.StreamHandler()
+    if stream is not None:
+        sh = logging.StreamHandler(stream)
         sh.setFormatter(formatter)
         lg.addHandler(sh)
 

docs/source/models/xtts.md

@@ -163,12 +163,13 @@ from TTS.api import TTS
 tts = TTS("tts_models/multilingual/multi-dataset/xtts_v2").to("cuda")
 
 # generate speech by cloning a voice using default settings
-tts.tts_to_file(text="It took me quite a long time to develop a voice, and now that I have it I'm not going to be silent.",
-                file_path="output.wav",
-                speaker="Ana Florence",
-                language="en",
-                split_sentences=True
-                )
+tts.tts_to_file(
+  text="It took me quite a long time to develop a voice, and now that I have it I'm not going to be silent.",
+  file_path="output.wav",
+  speaker="Ana Florence",
+  language="en",
+  split_sentences=True
+)
 ```
 
 
@@ -230,6 +231,11 @@ out = model.inference(
 torchaudio.save("xtts.wav", torch.tensor(out["wav"]).unsqueeze(0), 24000)
 ```
 
+You can also use the Coqui speakers:
+
+```python
+gpt_cond_latent, speaker_embedding = model.speaker_manager.speakers["Ana Florence"].values()
+```
 
 #### Streaming manually