online mix noise audio data in training step #2622
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,170 @@ | ||
| from __future__ import absolute_import, division, print_function | ||
|
|
||
| # Make sure we can import stuff from util/ | ||
| # This script needs to be run from the root of the DeepSpeech repository | ||
|
|
||
| from librosa import get_duration | ||
| from multiprocessing import Pool | ||
| from functools import partial | ||
| import math | ||
| import argparse | ||
| import sys | ||
| import os | ||
| import progressbar | ||
| sys.path.insert(1, os.path.join(sys.path[0], '..')) | ||
|
|
||
| from util.feeding import secs_to_hours | ||
|
|
||
| try: | ||
| from pydub import AudioSegment | ||
|
There was a problem hiding this comment. We want to keep dependencies at a minimum. Please check, if your required functionality couldn't be covered by e.g. |
||
| except ImportError as err: | ||
| print('[ImportError] try `sudo apt-get install ffmpeg && pip install pydub`') | ||
| raise err | ||
|
|
||
|
|
||
| def detect_silence(sound: AudioSegment, silence_threshold=-50.0, chunk_size=10): | ||
| start_trim = 0 # ms | ||
| sound_size = len(sound) | ||
| assert chunk_size > 0 # to avoid infinite loop | ||
| while sound[start_trim:(start_trim + chunk_size)].dBFS < silence_threshold and start_trim < sound_size: | ||
| start_trim += chunk_size | ||
|
|
||
| end_trim = sound_size | ||
| while sound[(end_trim - chunk_size):end_trim].dBFS < silence_threshold and end_trim > 0: | ||
| end_trim -= chunk_size | ||
|
|
||
| start_trim = min(sound_size, start_trim) | ||
| end_trim = max(0, end_trim) | ||
|
|
||
| return min([start_trim, end_trim]), max([start_trim, end_trim]) | ||
|
|
||
|
|
||
| def trim_silence_audio(sound: AudioSegment, silence_threshold=-50.0, chunk_size=10): | ||
| start_trim, end_trim = detect_silence(sound, silence_threshold, chunk_size) | ||
| return sound[start_trim:end_trim] | ||
|
|
||
|
|
||
| def convert(filename, dst_dirpath, dirpath, normalize, trim_silence, | ||
|
There was a problem hiding this comment. Please check, if how this is covered by or could be merged into the current audio.py. |
||
| min_duration_seconds, max_duration_seconds): | ||
| if not filename.endswith(('.wav', '.raw')): | ||
| return | ||
|
|
||
| filepath = os.path.join(dirpath, filename) | ||
| if filename.endswith('.wav'): | ||
| sound: AudioSegment = AudioSegment.from_file(filepath) | ||
| else: | ||
| try: | ||
| sound: AudioSegment = AudioSegment.from_raw(filepath, | ||
| sample_width=2, | ||
| frame_rate=44100, | ||
| channels=1) | ||
|
There was a problem hiding this comment. Please take |
||
| except Exception as err: # pylint: disable=broad-except | ||
| print('Retrying conversion: {}'.format(err)) | ||
|
|
||
| try: | ||
| sound: AudioSegment = AudioSegment.from_raw(filepath, | ||
| sample_width=2, | ||
| frame_rate=48000, | ||
| channels=1) | ||
| except Exception as err: # pylint: disable=broad-except | ||
| print('Skipping file {}, got error: {}'.format(filepath, err)) | ||
| return | ||
| try: | ||
| sound = sound.set_frame_rate(16000) | ||
|
There was a problem hiding this comment. Please make this command-line configurable. |
||
| except Exception as err: # pylint: disable=broad-except | ||
| print('Skipping {}'.format(err)) | ||
| return | ||
|
|
||
| n_splits = max(1, math.ceil(sound.duration_seconds / max_duration_seconds)) | ||
|
There was a problem hiding this comment. Great idea to split noise into chunks to limit wasted overlap during augmentation! |
||
| chunk_duration_ms = math.ceil(len(sound) / n_splits) | ||
| chunks = [] | ||
|
|
||
| for i in range(n_splits): | ||
| end_ms = min((i + 1) * chunk_duration_ms, len(sound)) | ||
| chunk = sound[(i * chunk_duration_ms):end_ms] | ||
| chunks.append(chunk) | ||
|
|
||
| for i, chunk in enumerate(chunks): | ||
| dst_path = os.path.join(dst_dirpath, str(i) + '_' + filename) | ||
| if dst_path.endswith('.raw'): | ||
| dst_path = dst_path[:-4] + '.wav' | ||
|
|
||
| if os.path.exists(dst_path): | ||
| print('Audio already exists: {}'.format(dst_path)) | ||
| return | ||
|
|
||
| if normalize: | ||
| chunk = chunk.normalize() | ||
| if chunk.dBFS < -30.0: | ||
| chunk = chunk.compress_dynamic_range().normalize() | ||
| if chunk.dBFS < -30.0: | ||
| chunk = chunk.compress_dynamic_range().normalize() | ||
| if trim_silence: | ||
| chunk = trim_silence_audio(chunk) | ||
|
|
||
| if chunk.duration_seconds < min_duration_seconds: | ||
| return | ||
| chunk.export(dst_path, format='wav') | ||
|
|
||
|
|
||
| def get_noise_duration(dst_dir): | ||
| duration = 0.0 | ||
| file_num = 0 | ||
| for dirpath, _, filenames in os.walk(dst_dir): | ||
| for f in filenames: | ||
| if not f.endswith('.wav'): | ||
| continue | ||
| duration += get_duration(filename=os.path.join(dirpath, f)) | ||
| file_num += 1 | ||
| return duration, file_num | ||
|
|
||
|
|
||
| def main(src_dir, | ||
| dst_dir, | ||
| min_duration_seconds, | ||
| max_duration_seconds, | ||
| normalize=True, | ||
| trim_silence=True): | ||
| assert os.path.exists(src_dir) | ||
| if not os.path.exists(dst_dir): | ||
| os.makedirs(dst_dir, exist_ok=False) | ||
| src_dir = os.path.abspath(src_dir) | ||
| dst_dir = os.path.abspath(dst_dir) | ||
|
|
||
| for dirpath, _, filenames in os.walk(src_dir): | ||
| dirpath = os.path.abspath(dirpath) | ||
| dst_dirpath = os.path.join( | ||
| dst_dir, dirpath.replace(src_dir, '').lstrip('/')) | ||
|
|
||
| print('Converting directory: {} -> {}'.format(dirpath, dst_dirpath)) | ||
| if not os.path.exists(dst_dirpath): | ||
| os.makedirs(dst_dirpath, exist_ok=False) | ||
|
|
||
| convert_func = partial(convert, | ||
| dst_dirpath=dst_dirpath, | ||
| dirpath=dirpath, | ||
| normalize=normalize, | ||
| trim_silence=trim_silence, | ||
| min_duration_seconds=min_duration_seconds, | ||
| max_duration_seconds=max_duration_seconds) | ||
|
|
||
| pool = Pool(processes=None) | ||
| pbar = progressbar.ProgressBar(prefix='Preparing Noise Dataset', max_value=len(filenames)).start() | ||
| for i, _ in enumerate(pool.imap_unordered(convert_func, filenames)): | ||
| pbar.update(i) | ||
| pbar.finish() | ||
|
|
||
|
|
||
| if __name__ == "__main__": | ||
| PARSER = argparse.ArgumentParser(description='Optimize noise files') | ||
| PARSER.add_argument('--from_dir', help='Convert wav from directory', type=str) | ||
| PARSER.add_argument('--to_dir', help='save wav to directory', type=str) | ||
|
There was a problem hiding this comment. This tool should also be able to produce SDBs like our SDB tool. I'll put up a PR for changing classes |
||
| PARSER.add_argument('--min_sec', help='min duration seconds of saved file', type=float, default=1.0) | ||
| PARSER.add_argument('--max_sec', help='max duration seconds of saved file', type=float, default=30.0) | ||
| PARSER.add_argument('--normalize', action='store_true', help='Normalize sound range, default is true', default=True) | ||
| PARSER.add_argument('--trim', action='store_true', help='Trim silence, default is true', default=True) | ||
| PARAMS = PARSER.parse_args() | ||
|
|
||
| main(PARAMS.from_dir, PARAMS.to_dir, PARAMS.min_sec, PARAMS.max_sec, PARAMS.normalize, PARAMS.trim) | ||
|
|
||
| DURATION, FILE_NUM = get_noise_duration(PARAMS.to_dir) | ||
| print("Your noise dataset has {} files and a duration of {}\n".format(FILE_NUM, secs_to_hours(DURATION))) | ||
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
How about
prepare_noise.py?