def train_model_for_phone(label_feat_pairs):
model = LinearRegression()
labels, gops = list(zip(*label_feat_pairs))
labels = np.array(labels).reshape(-1, 1)
gops = np.array(gops).reshape(-1, 1)
gops = PolynomialFeatures(2).fit_transform(gops)
gops, labels = balanced_sampling(gops, labels)
model.fit(gops, labels)
return model.coef_, model.intercept_
def train_model_for_phone(label_feat_pairs):
model = LinearRegression()
labels = []
gops = []
for label, gop in label_feat_pairs:
labels.append(label)
gops.append(gop)
labels = np.array(labels).reshape(-1, 1)
gops = np.array(gops).reshape(-1, 1)
gops = PolynomialFeatures(2).fit_transform(gops)
gops, labels = balanced_sampling(gops, labels)
model.fit(gops, labels)
return model.coef_, model.intercept_
def train_model_for_phone(label_feat_pairs):
model = RandomForestRegressor()
labels = []
feats = []
for label, feat in label_feat_pairs:
labels.append(label)
feats.append(feat[1:])
labels = np.array(labels).reshape(-1, 1)
feats = np.array(feats).reshape(-1, len(feats[0]))
feats, labels = balanced_sampling(feats, labels)
labels = labels.ravel()
model.fit(feats, labels)
return model
def main():
args = get_args()
# Phone symbol table
_, phone_int2sym = load_phone_symbol_table(args.phone_symbol_table)
# Human expert scores
score_of, phone_of = load_human_scores(args.human_scoring_json, floor=1)
# Prepare training data
train_data_of = {}
for key, gops in kaldi_io.read_post_scp(args.gop_scp):
for i, [(ph, gop)] in enumerate(gops):
ph_key = f'{key}.{i}'
if ph_key not in score_of:
print(f'Warning: no human score for {ph_key}')
continue
if phone_int2sym is not None and phone_int2sym[ph] != phone_of[
ph_key]:
print(f'Unmatch: {phone_int2sym[ph]} <--> {phone_of[ph_key]} ')
continue
score = score_of[ph_key]
if ph not in train_data_of:
train_data_of[ph] = []
train_data_of[ph].append((score, gop))
# Train polynomial regression
poly = PolynomialFeatures(2)
model_of = {}
for ph, pairs in train_data_of.items():
model = LinearRegression()
labels = []
gops = []
for label, gop in pairs:
labels.append(label)
gops.append(gop)
labels = np.array(labels).reshape(-1, 1)
gops = np.array(gops).reshape(-1, 1)
gops = poly.fit_transform(gops)
gops, labels = balanced_sampling(gops, labels)
model.fit(gops, labels)
model_of[ph] = (model.coef_, model.intercept_)
# Write to file
with open(args.model, 'wb') as f:
pickle.dump(model_of, f)
def main():
args = get_args()
# Phone symbol table
_, phone_int2sym = load_phone_symbol_table(args.phone_symbol_table)
# Human expert scores
score_of, phone_of = load_human_scores(args.human_scoring_json, floor=1)
# Prepare training data
train_data_of = {}
for ph_key, feat in kaldi_io.read_vec_flt_scp(args.feature_scp):
if ph_key not in score_of:
print(f'Warning: no human score for {ph_key}')
continue
if phone_int2sym is not None:
ph = int(feat[0])
if phone_int2sym[ph] != phone_of[ph_key]:
print(f'Unmatch: {phone_int2sym[ph]} <--> {phone_of[ph_key]} ')
continue
score = score_of[ph_key]
if ph not in train_data_of:
train_data_of[ph] = []
train_data_of[ph].append((score, feat))
# Train models
model_of = {}
for ph, pairs in train_data_of.items():
model = RandomForestRegressor()
labels = []
feats = []
for label, feat in pairs:
labels.append(label)
feats.append(feat[1:])
labels = np.array(labels).reshape(-1, 1)
feats = np.array(feats).reshape(-1, len(feats[0]))
feats, labels = balanced_sampling(feats, labels)
labels = labels.ravel()
model.fit(feats, labels)
model_of[ph] = model
print(f'Model of phone {ph} trained.')
# Write to file
with open(args.model, 'wb') as f:
pickle.dump(model_of, f)