Using speech2text for long duration of audio results in an error with GoogleAPI?

3 次查看（过去 30 天）
h612
h612 2019-8-17
编辑： Raja Palanimurugan 2019-8-27
I'm getting the folloing error when I try to convert speech to text for a longer duration:
'Sync input too long. For audio longer than 1 min use LongRunningRecognize with a ''uri'' parameter.'
I've modified my jason as
{
"key": "abc",
"uri":"https://speech.googleapis.com/v1/speech:longrunningrecognize"
}
But still the error is:
'Sync input too long. For audio longer than 1 min use LongRunningRecognize with a ''uri'' parameter.'
Here is my code:
transcriber = speechClient('Google');
[speech,SampleRate] = audioread('sound_meeting.wav');
text = speech2text(transcriber,speech(:,1),SampleRate,'HTTPTimeOut',25);

请先登录，再回答此问题。

回答（1 个）

Raja Palanimurugan
Raja Palanimurugan 2019-8-27
编辑：Raja Palanimurugan 2019-8-27
Transcribing long audio signals is not supported by the speech2text function yet. As a workaround, you can split the audio into smaller chunks before sending to speech2text. Hope that helps.

类别

在 Help Center 和 File Exchange 中查找有关 Audio I/O and Waveform Generation 的更多信息

标签

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by