I cant access anything in "Audio Content Creation", error "You don't have operation permissions"
I just created a speech service, but when I go to "Audio Content Creation", I can't do anything (New - Upload - Export) I tried to add myself as owner role, and other roles, but still, I can't do anything in Audio Content Creation.
Will Azure AI Speech generate styles such as "happy", "cheerful", "excited" automatically from the data given?
I've added data with about 750 utterances. 80% are normal sentences, while 10% are questions and the other 10% are exclamations. What will Speech Studio need to generate styles such as Happy, Cheerful, etc? Do I have to give it more data? Or will…
Bug Report: Mispronunciation of Welsh Contraction "i’w" in Azure Neural TTS
Subject: Bug Report: Mispronunciation of Welsh Contraction "i’w" in Azure Neural TTS Description: The Azure Neural TTS system is mispronouncing the Welsh contraction "i’w." Instead of producing the correct pronunciation…
Why am I getting a quota error?
I'm using Azure TTS and getting the following quota error: "You have reached the quota with your free-tier (F0) Speech resource. To continue to create audios with neural voices, switch to a standard paid resource, or upgrade your free-tier…
No module named 'azure' when using azure.cognitiveservices.speech
Hello, I have a problem with importing azure.cognitiveservices.speech. I pip install the package but when importing it I got this error. ModuleNotFoundError: No module named 'azure'
How to transcribe silences to train a custom STT model?
Hey! 🙂 I'm about to fine-tune a STT model with Audio + human-labeled transcript data. I've gone through the docs and I'm pretty confident that I've the right use case for this type of custom model training. Also, I already know how to organize the data…
Unable to delete audio file
Hi, I am using azure speech to text service. Originally i have video file and then getting audio file using ffmpeg. import azure.cognitiveservices.speech as speechsdk speech_config = speechsdk.SpeechConfig(subscription=key, endpoint=endpoint) …
Is it possible to recognize short words ("Yes", "No", "Ok") in speech sdk consistently
hello, I was experimenting with SPEECH SDK and I was using https://github.com/Azure-Samples/cognitive-services-speech-sdk/blob/master/quickstart/cpp/linux/from-microphone/helloworld.cpp I've adjusted code a bit, I was using the following…
In, e.g., 0001.sentence.json, quotation marks present in the original sentence are dropped, if that quotation mark occurs at the beginning or end of the detected sentence. Is this expected behavior?
This is mostly in the title. Initially, I suspected this was a bug in the JSON serialization since JSON also uses " to delimit its fields, and these also have to be escaped in SSML. Upon further investigation, however, i found it also affects…
![](https://techprofile.blob.core.windows.net/images/k3pGvNqV6k28ggaVHaTPrQ.png?8D803E)
Emotion Detection and Recognition from Text
What are some potential applications of Emotion Detection and Recognition from text, which aims to identify specific feelings like anger, disgust, fear, happiness, sadness, and surprise?
TTS繁體中文國語發音錯誤
「重考」發音應該是 ㄔㄨㄥˊ ㄎㄠˇ 「假期」發音應該是 ㄐㄧㄚˋ ㄑㄧˊ TTS 是收費服務,因此請儘快修正。 謝謝
Azure AI - Speech Studio - Error Message
Hi there, I receive this error message today. "为资源 xiaoshuoyuedu1 分配的角色尚未生效。 请让资源管理员配置__自定义子域__并启用 VNet 以使你的角色正常工作。" "The role assigned to resource xiaoshuoyuedu1 has not taken effect yet. Please have the resource administrator configure…
![](https://techprofile.blob.core.windows.net/images/k3pGvNqV6k28ggaVHaTPrQ.png?8D803E)
Speech recognition service is not working correctly
Hi, I'm using your speech service to recognize phrases spoken by a user in real time and evaluate their pronunciation. However, I am facing the following issues If I pass the reference text and set EnableMiscue =true, then all the wrong words the user…
Azure speech to text batch stucked on "Running" status and no percentage
this is the request: "azureRequest": { "displayName": "job_title...", "description": "job_title...", "locale": "it-it", "contentUrls": [ "{url of a wave…
Handling connection errors in Speech SDK
Hi, we are using Speech SDK (version 1.35.0, C++) for "speech to text". We use SpeechRecognizer->StartKeywordRecognitionAsync. While running the application, we lose connection sometimes and sometimes internet connection is okay, but we get…
![](https://techprofile.blob.core.windows.net/images/k3pGvNqV6k28ggaVHaTPrQ.png?8D803E)
Sample Data for different styles of Custom Neural Voices (happy, excited, sad).
I could find individual utterances for neutral speech, questions, and exclamations here: https://github.com/Azure-Samples/Cognitive-Speech-TTS/blob/master/CustomVoice/Sample%20Data/Individual%20utterances%20%2B%20matching%20script/SampleScript.txt To…
Do we need to close/suspend built-in AI voices (Ava, Andrew, Emma, Brian, etc) after using them to create a file in Audio Content Creation?
Hello, I understand that Custom Neural Voices need to be suspended after use due to their per-hour pricing. Do we also need to suspend anything after using Microsoft's built-in AI voices? I couldn't find specific information on this and want to avoid…
How to estimate the time needed to train a custom STT model?
Hey! I'm thinking about fine-tuning a STT model with Audio + human-labeled transcript data in Speech Studio. However, as I read through the docs, I can see that "If you switch to a base model that supports customization with audio data, the training…
How do you do pronunciation
Recently I had a script for a programming video, and I needed the word GUID, or goo id. I tried typing many different ways, and the only way I could get the word GUID, was to type goo hid, and use an audio editor and get rid of the H sound. Azure Speech…
training with mixed language in custom-stt(English & Korean)
Hi, I am working on training korean custom-stt, but in the training data , there are a few english words mixed in it. Some of them are well processed and accepted as train data but others get rejected such as winder, insulator, gripper, rewinding. below…
![](https://techprofile.blob.core.windows.net/images/BKeJHjfSr0Sh6sdaDxq_vg.png?8DC065)