Azure AI Speech
An Azure service that integrates speech processing into apps and services.
2,061 questions
Text to speech service return error code 1013
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(12.8, 68%, 11%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EJ%3C/text%3E%3C/svg%3E)
Joe
1
Reputation point
I have tried to call the TTS service by using SSML format but some of the text just returning 1013 error. It it because the text is too long?
CANCELED: ErrorDetails=Connection was closed by the remote host. Error code: 1013. Error details: Status(StatusCode="ResourceExhausted", Detail="the duration of the data sent in `mts.v1.InputChannel USP state: 4. Received audio size: 2399760 bytes.
Azure AI Speech
{count} votes
-
romungi-MSFT 48,906 Reputation points Microsoft Employee Moderator2022-10-20T10:01:10.703+00:00 @Joe Thanks for reporting the error. This error indicates a backend error with the service. In which region is your resource located?
-
Joe 1 Reputation point
2022-10-20T10:40:43.053+00:00 Hi Romungi, my service sits under
Southeast Asia. I think I figure out the issue becaus of the duration limit 10 minutes set for the TTS service. May I ask is there any suggestions for the long text TTS? (I wa using the SSML format) -
romungi-MSFT 48,906 Reputation points Microsoft Employee Moderator
2022-10-21T04:08:45.047+00:00 @Joe If you are seeing the limitation of 10 minutes then you are hitting the limit with the short audio API. You will have to use the long audio API to synthesize larger content. Please see the referenced page.
The URL will change to the following:
https://<region>.customvoice.api.speech.microsoft.com/api/texttospeech/v3.0/longaudiosynthesis -
Joe 1 Reputation point
2022-10-22T12:07:22.917+00:00 Thanks for the replying! I have used the long audio endpoint before but found the SSML is more flexible. So the SSML doesn't support any long text at this stage? The only way to make it work is using the long audio endpoint?
-
romungi-MSFT 48,906 Reputation points Microsoft Employee Moderator
2022-10-25T05:35:53.64+00:00 Yes, the long audio has some limited support for some SSML components like audio, mstts:backgroundaudio, and lexicon elements.
Sign in to comment
Sign in to answer