This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
![](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(259.20000000000005, 78%, 35%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EGT%3C/text%3E%3C/svg%3E)
Hello,
So we've started getting the following error: "HTTPError: 400 Client Error: Synthesis failed. StatusCode: NotFound, Details: service does not exist: service endpoint de-de-combogroup48neural-clusterip.tts-standardvoice-backend.svc.cluster.local:9001 failed dns validation."
We are using the voice fr-FR-RemyMultilingualNeural (we don't get the error if using en-US-JennyNeural or fr-FR-MauriceNeural for example) from the endpoint https://germanywestcentral.tts.speech.microsoft.com/cognitiveservices/v1
We've been using that voice from that endpoint for quite some time and the issues started to occured since 22 oct. 2024 around16:49 (UTC+01:00).
![](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(240, 23%, 33%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ERW%3C/text%3E%3C/svg%3E)
Same for de-DE-FlorianMultilingualNeural and de-DE-SeraphinaMultilingualNeural via endpoint https://germanywestcentral.tts.speech.microsoft.com/cognitiveservices/v1
Even when you try them on the console (custom text, not the pre-recorded audio), there is no audio output. delivered. I also tested various other MultilingualNeural voices and several had the same problem.
At test request would look like this (resulting in HTTP 400): $ curl -i -X "POST" "https://germanywestcentral.tts.speech.microsoft.com/cognitiveservices/v1" \ -H 'Ocp-Apim-Subscription-Key: XXX' \ -H 'X-Microsoft-OutputFormat: raw-8khz-8bit-mono-mulaw' \ -H 'Content-Type: application/ssml+xml' \ -H 'User-Agent: virtualq/tts' \ -d '<speak version=\'1.0\' xml:lang=\'en-US\'> <voice xml:lang=\'en-US\' name=\'de-DE-FlorianMultilingualNeural\'> Test </voice> </speak>'
0 commentsNo commentsReport a concern
![](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(6.4, 61%, 10%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ES%3C/text%3E%3C/svg%3E)
Hi @Gymglish Team,
We are reaching out to the internal team to get more information related to your query and will get back to you as soon as we have an update.
![](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(6.4, 53%, 10%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EMM%3C/text%3E%3C/svg%3E)
same for me "fr-FR-VivienneMultilingualNeural" just stopped working, the other voices work
Hi @madicke mbaye,
Sorry for the trouble you are facing. The tried to repro the issue without any issues using fr-FR-VivienneMultilingualNeural. Still if you face any issues, please let us know. However, I would recommend creating a new thread on the same forum with as much details about your issue as possible. That would make sure that your issue has better visibility in the community.
Thank you.
Hi @Gymglish Team,
We haven’t heard from you on the last response and was just checking back to see if you can able to use fr-FR-RemyMultilingualNeural. Thank you
Hi @Gymglish Team,
Thank you for your patience.
We’ve confirmed that the issue with fr-FR-RemyMultilingualNeural on the Germany West Central endpoint, which began on Oct 22, 2024, has been resolved. We could reproduce the scenario, and the voice is now working as expected. Please let us know if you experience further issues.