File size: 7,439 Bytes
cbbb8b2 5922aa4 cbbb8b2 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 |
#RECORDING INFORMATION
#
#The Romanian speech synthesis (RSS) corpus was recorded in a hemianechoic chamber (anechoic walls and ceiling; floor partially anechoic) at the University of Edinburgh. We used three high quality studio microphones: a Neumann u89i (large diaphragm condenser), a Sennheiser MKH 800 (small diaphragm condenser with very wide bandwidth) and a DPA 4035 (headset-mounted condenser). Although the current release includes only speech data recorded via Sennheiser MKH 800, we may release speech data recorded via other microphones in the future. All recordings were made at 96 kHz sampling frequency and 24 bits per sample, then downsampled to 48 kHz sampling frequency. For recording, downsampling and bit rate conversion, we used ProTools HD hardware and software. We conducted 8 sessions over the course of a month, recording about 500 sentences in each session. At the start of each session, the speaker listened to a previously recorded sample, in order to attain a similar voice quality and intonation.
#
#ACKNOWLEDGEMENTS
#
#The RSS database is constructed by:
#
# Adriana Stan (Technical University of Cluj-Napoca, Romania)
# Junichi Yamagishi (University of Edinburgh, United Kingdom)
# Simon King (University of Edinburgh, United Kingdom)
# Matthew Aylett (Cereproc)
#
#The following people and organisations have contributed to the development of the database in various ways. It is their work that makes it all possible.
#
# Korin Richmond
# Rob Clark
# Oliver Watts
# Chris Pidcock
# Graham Leary
# Blaise Potard
# Adevarul Online
# DEX Online - Catalin Francu
# Paul Borza
# Ovidiu Sabou
# Doina Tatar
# Mircea Giurgiu
# European Social Fund POSDRU/6/1.5/S/5
#
#and others too.
#
#
# This script for Hugging Face's datasets library was written by Théo Gigant
import csv
import json
import os
import datasets
_CITATION = """\
@article{Stan2011442,
author = {Adriana Stan and Junichi Yamagishi and Simon King and
Matthew Aylett},
title = {The {R}omanian speech synthesis ({RSS}) corpus:
Building a high quality {HMM}-based speech synthesis
system using a high sampling rate},
journal = {Speech Communication},
volume = {53},
number = {3},
pages = {442--450},
note = {},
abstract = {This paper first introduces a newly-recorded high
quality Romanian speech corpus designed for speech
synthesis, called ''RSS'', along with Romanian
front-end text processing modules and HMM-based
synthetic voices built from the corpus. All of these
are now freely available for academic use in order to
promote Romanian speech technology research. The RSS
corpus comprises 3500 training sentences and 500 test
sentences uttered by a female speaker and was recorded
using multiple microphones at 96 kHz sampling
frequency in a hemianechoic chamber. The details of the
new Romanian text processor we have developed are also
given. Using the database, we then revisit some basic
configuration choices of speech synthesis, such as
waveform sampling frequency and auditory frequency
warping scale, with the aim of improving speaker
similarity, which is an acknowledged weakness of
current HMM-based speech synthesisers. As we
demonstrate using perceptual tests, these configuration
choices can make substantial differences to the quality
of the synthetic speech. Contrary to common practice in
automatic speech recognition, higher waveform sampling
frequencies can offer enhanced feature extraction and
improved speaker similarity for HMM-based speech
synthesis.},
doi = {10.1016/j.specom.2010.12.002},
issn = {0167-6393},
keywords = {Speech synthesis, HTS, Romanian, HMMs, Sampling
frequency, Auditory scale},
url = {http://www.sciencedirect.com/science/article/pii/S0167639310002074},
year = 2011
}
"""
_DESCRIPTION = """\
The Romanian speech synthesis (RSS) corpus was recorded in a hemianechoic chamber (anechoic walls and ceiling; floor partially anechoic) at the University of Edinburgh. We used three high quality studio microphones: a Neumann u89i (large diaphragm condenser), a Sennheiser MKH 800 (small diaphragm condenser with very wide bandwidth) and a DPA 4035 (headset-mounted condenser). Although the current release includes only speech data recorded via Sennheiser MKH 800, we may release speech data recorded via other microphones in the future. All recordings were made at 96 kHz sampling frequency and 24 bits per sample, then downsampled to 48 kHz sampling frequency. For recording, downsampling and bit rate conversion, we used ProTools HD hardware and software. We conducted 8 sessions over the course of a month, recording about 500 sentences in each session. At the start of each session, the speaker listened to a previously recorded sample, in order to attain a similar voice quality and intonation.
"""
_HOMEPAGE = "http://romaniantts.com/rssdb/"
_LICENSE = "CCPL"
_URLS = {
"ro": "RomanianDB_v.0.8.1.tgz",
}
class RomanianSpeechSynthesis(datasets.GeneratorBasedBuilder):
VERSION = datasets.Version("0.8.1")
BUILDER_CONFIGS = [
datasets.BuilderConfig(name="ro", version=VERSION, description=""),
]
DEFAULT_CONFIG_NAME = "ro"
def _info(self):
features = datasets.Features(
{
"sentence": datasets.Value("string"),
"audio": datasets.features.Audio(sampling_rate=48_000),
}
)
return datasets.DatasetInfo(
description=_DESCRIPTION,
features=features,
homepage=_HOMEPAGE,
license=_LICENSE,
citation=_CITATION,
)
def _split_generators(self, dl_manager):
urls = _URLS[self.config.name]
data_dir = dl_manager.download_and_extract(urls)
return [
datasets.SplitGenerator(
name=datasets.Split.TRAIN,
gen_kwargs={
"datapath": data_dir,
"split": "training",
},
),
datasets.SplitGenerator(
name=datasets.Split.TEST,
gen_kwargs={
"datapath": data_dir,
"split": "testing"
},
),
]
def _generate_examples(self, datapath, split):
key = 0
audio_folder = "wav"
for folder in (os.listdir(os.path.join(datapath, split, audio_folder))):
with open(os.path.join(datapath, split, "text", folder+".txt")) as text_file:
for line in text_file.readlines():
i = line[:3]
filename = f"adr_{folder}_{i}.wav"
local_path = os.path.join(split, audio_folder, folder, filename)
yield key, {
"sentence": line[4:-1],
"audio" : os.path.join(datapath, local_path)
}
key += 1
|