| | |
|---|
| | | import numpy as np |
|---|
| | | import torchaudio |
|---|
| | | import torchaudio.compliance.kaldi as kaldi |
|---|
| | | import yaml |
|---|
| | | |
|---|
| | | |
|---|
| | | def get_parser(): |
|---|
| | |
|---|
| | | required=True, |
|---|
| | | type=str, |
|---|
| | | help="the path of wav scps", |
|---|
| | | ) |
|---|
| | | parser.add_argument( |
|---|
| | | "--config_file", |
|---|
| | | type=str, |
|---|
| | | help="the config file for computing cmvn", |
|---|
| | | ) |
|---|
| | | parser.add_argument( |
|---|
| | | "--idx", |
|---|
| | |
|---|
| | | parser = get_parser() |
|---|
| | | args = parser.parse_args() |
|---|
| | | |
|---|
| | | wav_scp_file = os.path.join(args.wav_path + "{}/wav.scp".format(args.idx)) |
|---|
| | | cmvn_file = os.path.join(args.wav_path + "{}/cmvn.json".format(args.idx)) |
|---|
| | | wav_scp_file = os.path.join(args.wav_path, "wav.{}.scp".format(args.idx)) |
|---|
| | | cmvn_file = os.path.join(args.wav_path, "cmvn.{}.json".format(args.idx)) |
|---|
| | | |
|---|
| | | mean_stats = np.zeros(args.dims) |
|---|
| | | var_stats = np.zeros(args.dims) |
|---|
| | | mean_stats = np.zeros(args.dim) |
|---|
| | | var_stats = np.zeros(args.dim) |
|---|
| | | total_frames = 0 |
|---|
| | | |
|---|
| | | # with ReadHelper('ark:{}'.format(ark_file)) as ark_reader: |
|---|
| | |
|---|
| | | # mean_stats += np.sum(mat, axis=0) |
|---|
| | | # var_stats += np.sum(np.square(mat), axis=0) |
|---|
| | | # total_frames += mat.shape[0] |
|---|
| | | |
|---|
| | | with open(args.config_file) as f: |
|---|
| | | configs = yaml.safe_load(f) |
|---|
| | | frontend_configs = configs.get("frontend_conf", {}) |
|---|
| | | num_mel_bins = frontend_configs.get("n_mels", 80) |
|---|
| | | frame_length = frontend_configs.get("frame_length", 25) |
|---|
| | | frame_shift = frontend_configs.get("frame_shift", 10) |
|---|
| | | window_type = frontend_configs.get("window", "hamming") |
|---|
| | | resample_rate = frontend_configs.get("fs", 16000) |
|---|
| | | assert num_mel_bins == args.dim |
|---|
| | | |
|---|
| | | with open(wav_scp_file) as f: |
|---|
| | | lines = f.readlines() |
|---|
| | | for line in lines: |
|---|
| | | _, wav_file = line.strip().split() |
|---|
| | | fbank = compute_fbank(wav_file, num_mel_bins=args.dims) |
|---|
| | | fbank = compute_fbank(wav_file, |
|---|
| | | num_mel_bins=args.dim, |
|---|
| | | frame_length=frame_length, |
|---|
| | | frame_shift=frame_shift, |
|---|
| | | resample_rate=resample_rate, |
|---|
| | | window_type=window_type) |
|---|
| | | mean_stats += np.sum(fbank, axis=0) |
|---|
| | | var_stats += np.sum(np.square(fbank), axis=0) |
|---|
| | | total_frames += fbank.shape[0] |
|---|
