数据堂—343人西班牙语手机采集语音数据_引导

西班牙语手机采集语音数据_引导是由343名西班牙、墨西哥、阿根廷等国的西班牙语母语人员参与录制，每人50句，共计9.9小时，男女比例均匀；录音环境安静；所有文本由人工转写，准确率高；匹配主流苹果、安卓手机；西班牙语手机采集语音数据_引导可应用于语音识别，机器翻译，声纹识别等领域

Guided Spanish Speech Dataset Collected via Mobile Phones: This dataset was recorded by 343 native Spanish speakers from countries such as Spain, Mexico, Argentina and other Spanish-speaking regions. Each participant provided 50 utterances, resulting in a total of 9.9 hours of audio data with a balanced gender distribution. All audio recordings were conducted in quiet environments. All transcriptions were manually performed with high accuracy. The dataset is compatible with mainstream Apple and Android mobile devices, and can be applied to fields including speech recognition, machine translation, and speaker verification.

搜集汇总

背景与挑战

背景概述

该数据集包含343名母语西班牙语人士通过手机录制的语音数据，总时长9.9小时，采样率为16kHz，格式为WAV，用于西班牙语语音识别模型的测试。数据在安静室内环境下采集，涵盖多种场景，如车载和智能家居，并包含高精度人工转录文本。