Azure AI services
A group of Azure services, SDKs, and APIs designed to make apps more intelligent, engaging, and discoverable.
3,409 questions
How to use a custom translator in real-time voice translation
![](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(99.2, 99%, 19%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EA%3C/text%3E%3C/svg%3E)
集团 语言桥
0
Reputation points
I hope to be able to set a custom translation model in an API for real-time voice translation.
https://github.com/Azure-Samples/cognitive-services-speech-sdk/issues/2691
It is mentioned in the official GitHub issue that custom translation models can be set in real-time voice translation.
I used the categoryId here, but I always get an error: SessionId:3702ffc412d74365965fe01a65e12e47 ResultId:ba435d1d3703433ea7d83a7da8c85330 CancellationReason:Error CancellationErrorCode:BadRequest Error details:Connection was closed by the remote host. Error code: 1007. Error details:Translation custom models require CognitiveServices / AIServices resources. SessionId: 3702ffc412d74365965fe01a65e12e47
I want to confirm whether the method I am using is correct, and whether the categoryId I have set here is the Translator API Category ID? If not, could you please explain the correct usage, as I did not find this usage example in the official SDK examples.
Azure AI services
{count} votes
-
![](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(220.8, 32%, 31%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ERS%3C/text%3E%3C/svg%3E)
Ravada Shivaprasad 40 Reputation points Microsoft External Staff2025-04-17T16:36:35.8666667+00:00 Hi 集团 语言桥
To implement custom translation in real-time voice translation, you need a Cognitive Services/AI Services resource with both Speech and Custom Translation services enabled. The Category ID should be generated from your custom translation model in Azure AI Custom Translator.
As you are using the correct Category ID when making requests. If you encounter errors, check your service configurations, including endpoints and keys. The Speech SDK or Speech CLI can be used to access source transcriptions and translation outputs. Meanwhile prices may vary based on target languages and character counts.
Reference: Speech Translation,Custom Translator ,Azure Pricing
Thanks
-
Ravada Shivaprasad 40 Reputation points Microsoft External Staff
2025-04-18T18:08:11.5133333+00:00 -
Ravada Shivaprasad 40 Reputation points Microsoft External Staff
2025-04-21T05:33:07.4233333+00:00 -
集团 语言桥 0 Reputation points
2025-05-06T08:09:58.4533333+00:00 Sorry for taking so long to reply, a solution has been found before, what you said may not be very perfect, and it is not stated in the official documentation. Yes, you need to configure an AI services multi-service resource in a resource group, and enable the recognition and translation services at the same time, and use the key of the AI services multi-service to use the custom translation model normally
-
集团 语言桥 0 Reputation points
2025-05-06T08:16:35.65+00:00 - Turn on Speech Recognition and translation services at the same time
- You also need to enable AI services multi-service in the same resource group
- When using a key, the Speech Recognition service requires the key provided in the AI services multi-service
Sign in to comment
Sign in to answer