add speaker name
Browse files- libriheavy.py +7 -0
libriheavy.py
CHANGED
@@ -2,10 +2,12 @@ import json
|
|
2 |
import gzip
|
3 |
import os
|
4 |
from pathlib import Path
|
|
|
5 |
|
6 |
import datasets
|
7 |
import numpy as np
|
8 |
from tqdm import tqdm
|
|
|
9 |
|
10 |
logger = datasets.logging.get_logger(__name__)
|
11 |
|
@@ -117,6 +119,9 @@ class Libriheavy(datasets.GeneratorBasedBuilder):
|
|
117 |
metadata_path = dl_manager.download_and_extract(metadata_path)
|
118 |
with open(metadata_path, "r") as f:
|
119 |
speaker_metadata[speaker_id] = json.load(f)
|
|
|
|
|
|
|
120 |
with open(f"{metadata_cache}/{speaker_id}.json", "w") as f:
|
121 |
json.dump(speaker_metadata[speaker_id], f)
|
122 |
|
@@ -127,6 +132,7 @@ class Libriheavy(datasets.GeneratorBasedBuilder):
|
|
127 |
for chunk_id, chunk in metadata["chunks"].items():
|
128 |
chunk_dict = {
|
129 |
"speaker_id": speaker_id,
|
|
|
130 |
"id": f"{speaker_id}_{chunk_id}",
|
131 |
"audio": dl_manager.download(f"{PATH}/{speaker_id}/{chunk['npz'].replace('.gz', '')}"),
|
132 |
"text": dl_manager.download(f"{PATH}/{speaker_id}/{chunk['json']}"),
|
@@ -174,6 +180,7 @@ class Libriheavy(datasets.GeneratorBasedBuilder):
|
|
174 |
result = {
|
175 |
"id": chunk["speaker_id"] + "_" + utterance_id,
|
176 |
"speaker_id": chunk["speaker_id"],
|
|
|
177 |
"speaker_vec": npz_item["d_vector"][0],
|
178 |
"audio": chunk["audio"],
|
179 |
"text": chunk["text"],
|
|
|
2 |
import gzip
|
3 |
import os
|
4 |
from pathlib import Path
|
5 |
+
import re
|
6 |
|
7 |
import datasets
|
8 |
import numpy as np
|
9 |
from tqdm import tqdm
|
10 |
+
import requests
|
11 |
|
12 |
logger = datasets.logging.get_logger(__name__)
|
13 |
|
|
|
119 |
metadata_path = dl_manager.download_and_extract(metadata_path)
|
120 |
with open(metadata_path, "r") as f:
|
121 |
speaker_metadata[speaker_id] = json.load(f)
|
122 |
+
speaker_name = requests.get(f"https://librivox.org/reader/{speaker_id}").text
|
123 |
+
speaker_name = re.findall("<h1>([^<>]+)</h1>", speaker_name)[0]
|
124 |
+
speaker_metadata[speaker_id]["name"] = speaker_name
|
125 |
with open(f"{metadata_cache}/{speaker_id}.json", "w") as f:
|
126 |
json.dump(speaker_metadata[speaker_id], f)
|
127 |
|
|
|
132 |
for chunk_id, chunk in metadata["chunks"].items():
|
133 |
chunk_dict = {
|
134 |
"speaker_id": speaker_id,
|
135 |
+
"speaker_name": metadata["name"],
|
136 |
"id": f"{speaker_id}_{chunk_id}",
|
137 |
"audio": dl_manager.download(f"{PATH}/{speaker_id}/{chunk['npz'].replace('.gz', '')}"),
|
138 |
"text": dl_manager.download(f"{PATH}/{speaker_id}/{chunk['json']}"),
|
|
|
180 |
result = {
|
181 |
"id": chunk["speaker_id"] + "_" + utterance_id,
|
182 |
"speaker_id": chunk["speaker_id"],
|
183 |
+
"speaker_name": chunk["speaker_name"],
|
184 |
"speaker_vec": npz_item["d_vector"][0],
|
185 |
"audio": chunk["audio"],
|
186 |
"text": chunk["text"],
|