here i cannot find To create a custom avatar endpoint, follow these steps: Sign in to Speech Studio. Navigate to Custom Avatar > Your project name > Train model.
i cannot find custom avatar key after sign in to the speech studio .
Inquiry Regarding Azure AI Speech Error
Dear Azure Support Team I recently encountered an issue while using Azure AI Speech service with recordings from the VoiceMemo app on iPhone. Specifically, when attempting to process recordings of approximately 30 minutes in length, I received the…
Speech Studio Audio Content Creation (x) Content Format and Audio Export Fail
I discovered https://speech.microsoft.com/portal, audio creation tile. (I think it should be the first one and described as "interactive batch TTS web interface.") I uploaded a file named test.txt, which has two paragraphs. For decades now,…
400 Bad request using whisper with AzureCliCredentials
I'm trying to use Whisper using the AzureCliCredential and i always get an error as follow { code: 'Request is badly formated', message: 'Resource Id is badly formed: NA' } my very simple code is : import * as fs from "fs"; import {…
How to transcribe foreign names and words within English sentences
I use Azure Speech to transcribe audio files in English through a Java application. There are however some reoccuring foreign words and names (Arabic) used in the middle of the English sentences and these are not properly transcribed. What is the…
Markdown to SSML ?
Does anyone know of a basic "preparer-converter" that takes a markdown (.md) file and converts it into an SSML file?
import user_config_helper
not able to find library to import user_config_helper import user_config_helper
Regarding usage cost calculation using Azure Retail Price API
We are using Azure subscription with the Standard Tier. We have a requirement to calculate the monthly usage cost in JPY (Japanese Yen) of the Azure Speech to Text service and Azure Blob Storage in our application. we analyzed the Azure Retail Price API…
Multilingual voice mispronounces Ukrainian as Russian
How can I resolve the issue of multilingual voices pronouncing Ukrainian as Russian when using Text to Speech with the Microsoft.CognitiveServices.Speech package in C#? Explicitly specifying the language in the code through the SpeechSynthesisLanguage…
Issue with speech-to-text service
While converting the given wave file from Speech-to-Text using Microsoft's Speech-to-Text service, it is not detecting "No" at 57th second in this file but detecting at 1:12 min and in other places. Speech recognised is as follow RECOGNIZED:…
Cognitive services pronunciation assessment always gives 100% score, even with badly pronounced words
I built a svelte (javascript) application that uses the microsoft speech sdk (v1.36), and i am using it to evaluate pronunciation in 3 languages: english, german and french. Initially i was using RecognizeOnceAsync() which waits for silence at the end of…
Batch TTS with REST: YourSynthesisId and other intro questions
I got the REST API to work on macos. Yeah!!! I could hear the output from the sample code. Alas, now I would like to submit a longer document I wrote to batch TTS and post it as my podcast. I am taking the example right off the webpage, and just…
Can you add a phrase list to the CallMediaRecognizeSpeechOptions class when using speech-to- text cognitive services from azure communications service
I am using ACS to access a multi-service Cognitive Services endpoint and doing recognition of speech input in real time via acs/telephone. I am using the default model provided by Microsoft. This is sufficient in most case but I have some place names…
![](https://techprofile.blob.core.windows.net/images/k3pGvNqV6k28ggaVHaTPrQ.png?8D803E)
Is it possible to specify in Speech SDK to always use "lbs" instead of "£" when "pounds" is recognized?
Hi, is it possible somehow to configure speech sdk in a way when word "pound" is detected that it is always meant to be lbs, not £, for example when I say, "99 pounds" it is detected as "99 lbs", but if I said, "100…
Set sound threshold for microsoft speech-to-text
Hi, It is possible setting a volume-threshold for the speech that gets transcribed? Such that if the speech is below a certain threshold then it would not get transcribed. I am using the speechSDK Br, Daniel
macos cli starter guide
I am trying to play around with azure text to speech on macos. the instructions are woefully incomplete. I start with…
Why my TTS is suddenly become bad? Speed & punctuation isn't working properly.
This morning I tried to work on my TTS file using Brian's voice. But once I listened to the speech, the punctuation & speed weren't working properly. Also, it seems that his voice became monotone. I've tried with an already-finished project to see if…
![](https://techprofile.blob.core.windows.net/images/KhnRGP5_AwAAAAAAAAAAAA.png?8DBA61)
ServiceTimeOut Errors in Azure Speech services Real time Transcription
Hello Team, Kindly need your help on the issue facing while using real time transcription with custom domain in azure Speech service with ASR Engines. Please find the below error message. ERROR: Cancelled: ErrorCode=ServiceTimeout,ErrorDetails=Due to…
can I be part of Seeing AI project in any way & get acknowledged for any contributions towards it
Hi Team, I found the "seeing AI" project quite interesting. Can I be part of Seeing AI project in any way & get acknowledged for any contributions towards it. By reading the "critical" comments about the app in Google play store I…
What is the name of the model that Azure Speech uses for Hebrew?
I'm using Azure Speech to transcribe Hebrew audio, but I'm not sure which model it uses. Can someone please provide this information? Is it the whisper model or something else? I'm getting better results than the whisper model with openai, so I'd like to…
![](https://techprofile.blob.core.windows.net/images/DFUdJReiskW7xaWenka5lQ.png?8DAE8D)