experiments
/
STT
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Undo late-imports

This commit is contained in:
Reuben Morais 2021-12-03 15:43:42 +01:00
parent 479d963155
commit ff24a8b917
1 changed files with 89 additions and 95 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

184
training/coqui_stt_training/transcribe.py
View File

@ -1,12 +1,5 @@
#!/usr/bin/env python #!/usr/bin/env python
# -*- coding: utf-8 -*- # -*- coding: utf-8 -*-
# This script is structured in a weird way, with delayed imports. This is due
# to the use of multiprocessing. TensorFlow cannot handle forking, and even with
# the spawn strategy set to "spawn" it still leads to weird problems, so we
# restructure the code so that TensorFlow is only imported inside the child
# processes.
import glob import glob
import itertools import itertools
import json import json
@ -18,28 +11,31 @@ from multiprocessing import Pool, Lock, cpu_count
from pathlib import Path from pathlib import Path
from typing import Optional, List, Tuple from typing import Optional, List, Tuple
LOG_LEVEL_INDEX = sys.argv.index("--log_level") + 1 if "--log_level" in sys.argv else 0
DESIRED_LOG_LEVEL = (
sys.argv[LOG_LEVEL_INDEX] if 0 < LOG_LEVEL_INDEX < len(sys.argv) else "3"
)
os.environ["TF_CPP_MIN_LOG_LEVEL"] = DESIRED_LOG_LEVEL
# Hide GPUs to prevent issues with child processes trying to use the same GPU
os.environ["CUDA_VISIBLE_DEVICES"] = "-1"
import tensorflow as tf
import tensorflow.compat.v1 as tfv1
from coqui_stt_ctcdecoder import Scorer, ctc_beam_search_decoder_batch from coqui_stt_ctcdecoder import Scorer, ctc_beam_search_decoder_batch
from coqui_stt_training.train import create_model
from coqui_stt_training.util.audio import AudioFile
from coqui_stt_training.util.checkpoints import load_graph_for_evaluation
from coqui_stt_training.util.config import (
BaseSttConfig,
Config,
initialize_globals_from_instance,
)
from coqui_stt_training.util.feeding import split_audio_file
from coqui_stt_training.util.helpers import check_ctcdecoder_version
from tqdm import tqdm from tqdm import tqdm
def transcribe_file(audio_path: Path, tlog_path: Path): def transcribe_file(audio_path: Path, tlog_path: Path):
log_level_index = (
sys.argv.index("--log_level") + 1 if "--log_level" in sys.argv else 0
)
desired_log_level = (
sys.argv[log_level_index] if 0 < log_level_index < len(sys.argv) else "3"
)
os.environ["TF_CPP_MIN_LOG_LEVEL"] = desired_log_level
import tensorflow as tf
import tensorflow.compat.v1 as tfv1
from coqui_stt_training.train import create_model
from coqui_stt_training.util.audio import AudioFile
from coqui_stt_training.util.checkpoints import load_graph_for_evaluation
from coqui_stt_training.util.config import Config
from coqui_stt_training.util.feeding import split_audio_file
initialize_transcribe_config() initialize_transcribe_config()
scorer = None scorer = None
@ -113,8 +109,6 @@ def step_function(job):
def transcribe_many(src_paths, dst_paths): def transcribe_many(src_paths, dst_paths):
from coqui_stt_training.util.config import Config, log_progress
# Create list of items to be processed: [(i, src_path[i], dst_paths[i])] # Create list of items to be processed: [(i, src_path[i], dst_paths[i])]
jobs = zip(itertools.count(), src_paths, dst_paths) jobs = zip(itertools.count(), src_paths, dst_paths)
@ -134,11 +128,17 @@ def transcribe_many(src_paths, dst_paths):
cwd = Path.cwd() cwd = Path.cwd()
for result in process_iterable: for result in process_iterable:
idx, src, dst = result idx, src, dst = result
# Revert to relative to avoid spamming logs # Revert to relative if possible to make logs more concise
if not src.is_absolute(): # if path is not relative to cwd, use the absolute path
# (Path.is_relative_to is only available in Python >=3.9)
try:
src = src.relative_to(cwd) src = src.relative_to(cwd)
if not dst.is_absolute(): except ValueError:
pass
try:
dst = dst.relative_to(cwd) dst = dst.relative_to(cwd)
except ValueError:
pass
tqdm.write(f'[{idx+1}]: "{src}" -> "{dst}"') tqdm.write(f'[{idx+1}]: "{src}" -> "{dst}"')
@ -187,14 +187,12 @@ def get_tasks_from_dir(src_dir: Path, recursive: bool) -> Tuple[List[Path], List
transcription results. transcription results.
""" """
glob_method = src_dir.rglob if recursive else src_dir.glob glob_method = src_dir.rglob if recursive else src_dir.glob
src_paths = list(itertools.chain(glob_method("*.wav"), glob_method("*.opus"))) src_paths = list(glob_method("*.wav"))
dst_paths = [path.with_suffix(".tlog") for path in src_paths] dst_paths = [path.with_suffix(".tlog") for path in src_paths]
return src_paths, dst_paths return src_paths, dst_paths
def transcribe(): def transcribe():
from coqui_stt_training.util.config import Config
initialize_transcribe_config() initialize_transcribe_config()
src_path = Path(Config.src).resolve() src_path = Path(Config.src).resolve()
@ -236,89 +234,85 @@ def transcribe():
transcribe_many(src_paths, dst_paths) transcribe_many(src_paths, dst_paths)
def initialize_transcribe_config(): @dataclass
from coqui_stt_training.util.config import ( class TranscribeConfig(BaseSttConfig):
BaseSttConfig, src: str = field(
initialize_globals_from_instance, default="",
metadata=dict(
help="Source path to an audio file or directory or catalog file. "
"Catalog files should be formatted from DSAlign. A directory "
"will be recursively searched for audio. If --dst not set, "
"transcription logs (.tlog) will be written in-place using the "
'source filenames with suffix ".tlog" instead of the original.'
),
) )
@dataclass dst: str = field(
class TranscribeConfig(BaseSttConfig): default="",
src: str = field( metadata=dict(
default="", help="path for writing the transcription log or logs (.tlog). "
metadata=dict( "If --src is a directory, this one also has to be a directory "
help="Source path to an audio file or directory or catalog file. " "and the required sub-dir tree of --src will get replicated."
"Catalog files should be formatted from DSAlign. A directory " ),
"will be recursively searched for audio. If --dst not set, " )
"transcription logs (.tlog) will be written in-place using the "
'source filenames with suffix ".tlog" instead of the original.'
),
)
dst: str = field( recursive: bool = field(
default="", default=False,
metadata=dict( metadata=dict(help="scan source directory recursively for audio"),
help="path for writing the transcription log or logs (.tlog). " )
"If --src is a directory, this one also has to be a directory "
"and the required sub-dir tree of --src will get replicated."
),
)
recursive: bool = field( force: bool = field(
default=False, default=False,
metadata=dict(help="scan source directory recursively for audio"), metadata=dict(
) help="Forces re-transcribing and overwriting of already existing "
"transcription logs (.tlog)"
),
)
force: bool = field( vad_aggressiveness: int = field(
default=False, default=3,
metadata=dict( metadata=dict(help="VAD aggressiveness setting (0=lowest, 3=highest)"),
help="Forces re-transcribing and overwriting of already existing " )
"transcription logs (.tlog)"
),
)
vad_aggressiveness: int = field( batch_size: int = field(
default=3, default=40,
metadata=dict(help="VAD aggressiveness setting (0=lowest, 3=highest)"), metadata=dict(help="Default batch size"),
) )
batch_size: int = field( outlier_duration_ms: int = field(
default=40, default=10000,
metadata=dict(help="Default batch size"), metadata=dict(
) help="Duration in ms after which samples are considered outliers"
),
)
outlier_duration_ms: int = field( outlier_batch_size: int = field(
default=10000, default=1,
metadata=dict( metadata=dict(help="Batch size for duration outliers (defaults to 1)"),
help="Duration in ms after which samples are considered outliers" )
),
)
outlier_batch_size: int = field( def __post_init__(self):
default=1, if os.path.isfile(self.src) and self.src.endswith(".catalog") and self.dst:
metadata=dict(help="Batch size for duration outliers (defaults to 1)"), raise RuntimeError(
) "Parameter --dst not supported if --src points to a catalog"
)
def __post_init__(self): if os.path.isdir(self.src):
if os.path.isfile(self.src) and self.src.endswith(".catalog") and self.dst: if self.dst:
raise RuntimeError( raise RuntimeError(
"Parameter --dst not supported if --src points to a catalog" "Destination path not supported for batch decoding jobs."
) )
if os.path.isdir(self.src): super().__post_init__()
if self.dst:
raise RuntimeError(
"Destination path not supported for batch decoding jobs."
)
super().__post_init__()
def initialize_transcribe_config():
config = TranscribeConfig.init_from_argparse(arg_prefix="") config = TranscribeConfig.init_from_argparse(arg_prefix="")
initialize_globals_from_instance(config) initialize_globals_from_instance(config)
def main(): def main():
from coqui_stt_training.util.helpers import check_ctcdecoder_version assert not tf.test.is_gpu_available()
# Set start method to spawn on all platforms to avoid issues with TensorFlow # Set start method to spawn on all platforms to avoid issues with TensorFlow
multiprocessing.set_start_method("spawn") multiprocessing.set_start_method("spawn")