AWS Transcribe输入文件的最短持续时间为1秒。以下是使用AWS SDK for Python(Boto3)将输入文件传递给AWS Transcribe的代码示例:
import boto3
# 创建一个transcribe客户端
transcribe = boto3.client('transcribe')
# 指定转录作业名称和输入文件位置
job_name = "example-job"
input_location = "s3://example-bucket/input.mp3"
# 检查输入文件的长度是否超过1秒
media_info = transcribe.get_media_info(
Media = {
'MediaFileUri': input_location
}
)
if media_info['MediaInfo']['MediaDurationInSeconds'] < 1:
raise ValueError("输入文件持续时间不能低于1秒")
# 开始转录作业
transcribe.start_transcription_job(
TranscriptionJobName = job_name,
Media = {
'MediaFileUri': input_location
},
MediaFormat = "mp3",
LanguageCode = "en-US"
)
# 等待转录作业完成
while True:
status = transcribe.get_transcription_job(TranscriptionJobName = job_name)
if status['TranscriptionJob']['TranscriptionJobStatus'] in ['COMPLETED', 'FAILED']:
break