Azure AI Speech
An Azure service that integrates speech processing into apps and services.
1,836 questions
0 answers
Request for Access to Azure Speaker Recognition Service
Dear Azure Support Team, I hope this message finds you well. My name is Jamil Al Nashash, and I am currently working on a project that involves speech to text. , I am looking to integrate Speaker Recognition into our application using Azure speech AI…
asked 2024-12-26T18:46:21.97+00:00
![](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(233.6, 54%, 32%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EJA%3C/text%3E%3C/svg%3E)
Jamil Al Nashash
0
Reputation points
edited the question 2024-12-27T20:56:00.5166667+00:00
Grmacjon-MSFT
18,646
Reputation points
0 answers
Questions about AI Speech test (evaluation) error
questions regarding the abnormal behavior in Azure AI Speech. 1. Model testing failed (Invalid or empty input data for evaluation) When I tried to create a test with the uploaded dataset as a sample data, I ran into the model testing failed with invalid…
asked 2024-12-24T00:26:52.62+00:00
![](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(310.4, 41%, 40%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EHC%3C/text%3E%3C/svg%3E)
Hyo Choi
0
Reputation points Microsoft Employee
commented 2024-12-27T01:14:31.3566667+00:00
![](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(16, 23%, 11%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EAD%3C/text%3E%3C/svg%3E)
Avinash Devarakonda
605
Reputation points Microsoft Vendor
1 answer
One of the answers was accepted by the question author.
What input formats does Batch Speech to Text support?
Are there any documentation or guidelines regarding the input formats supported by the Batch Speech to Text service? I have two mp4 files with different properties; one can be transcribed (bitrate 62kbps, mono, 16000kHz) , while the other cannot…
asked 2024-12-19T03:13:30.4533333+00:00
![](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(137.6, 91%, 23%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ES0%3C/text%3E%3C/svg%3E)
日立s 018
20
Reputation points
commented 2024-12-26T10:33:15.1433333+00:00
![](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(198.4, 14.000000000000002%, 29%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ETV%3C/text%3E%3C/svg%3E)
Tong Viet Anh
0
Reputation points
0 answers
Status: 500 trying to upload a large 400mb file to Video Translation service in Speech Studio
See attached image, works with small files but times out on large file 400mb, 40 min.
asked 2024-12-21T16:07:27.9533333+00:00
![](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(243.2, 20%, 33%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EV%3C/text%3E%3C/svg%3E)
VK
0
Reputation points
commented 2024-12-26T04:21:37.2966667+00:00
Avinash Devarakonda
605
Reputation points Microsoft Vendor
1 answer
"SSML parsing error: 0x8004801c - Wanted data exists" fails with lexicon based on 'fr-CA-SylvieNeural' voice in Azure Speech SDK, but works via Web Portal"
Hello, We are using the microsoft-cognitiveservices-speech-sdk npm module for text-to-speech synthesis in our Node.js application. We have an SSML segment as follows: <?xml version="1.0"?> <speak version="1.0"…
asked 2024-09-19T23:04:40.2766667+00:00
![](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(185.6, 28.999999999999996%, 27%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EHM%3C/text%3E%3C/svg%3E)
Hugo Machefer
0
Reputation points
commented 2024-12-26T03:21:09.1166667+00:00
![](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(198.4, 80%, 29%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EHM%3C/text%3E%3C/svg%3E)
Hugo MACHEFER
0
Reputation points
1 answer
Speechsynthesizer causes JVM crash during finalization (Speech client-sdk for Java)
https://github.com/Azure-Samples/cognitive-services-speech-sdk/issues/2701 As I detailed in the above github issue, there is a reproducible bug in the Java client-sdk for Azure Speech. During Finalization of Speechsynthesizer Instance it can cause a…
asked 2024-12-17T12:41:19.6966667+00:00
![](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(73.60000000000001, 86%, 17%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ELM%3C/text%3E%3C/svg%3E)
Lucas Mikulla
0
Reputation points
commented 2024-12-24T19:55:31.3866667+00:00
Lucas Mikulla
0
Reputation points
2 answers
Inquiry About Azure Speech SDK for Apple Vision OS
Dear Azure Support TeamWe have encountered some challenges and would appreciate your assistance. Here are the details of our inquiry: Current Status: We are currently trying to implement the Azure Speech SDK on Apple Vision OS. Unfortunately, we…
asked 2024-12-16T09:34:24.6666667+00:00
![](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(54.400000000000006, 22%, 15%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EXB%3C/text%3E%3C/svg%3E)
XRSPACE-Bowen Huang
0
Reputation points
commented 2024-12-24T09:58:41.43+00:00
![](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(281.6, 88%, 37%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ESA%3C/text%3E%3C/svg%3E)
Saideep Anchuri
685
Reputation points Microsoft Vendor
1 answer
Accuracy issues reading numbers with Text to speech voices
I wanted your perspective on an issue im having with 2 new voices reading numbers, occasionally Adam Multilingual and Lewis Multilingual will read numbers that start with "four" in a way that sounds like "for five six seven" like…
asked 2024-12-20T19:23:41.6233333+00:00
![](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(275.2, 54%, 36%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EAS%3C/text%3E%3C/svg%3E)
Andrew Silagy
0
Reputation points
commented 2024-12-24T06:59:39.34+00:00
![](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(153.6, 24%, 24%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ENM%3C/text%3E%3C/svg%3E)
navba-MSFT
26,970
Reputation points Microsoft Employee
2 answers
I would like to know if there are any other avatars besides the Asian figures for text to speech ? How do I access them?
I would like to have the option to selfct differen avatar figures besides the Asian one shown.
asked 2024-11-16T01:08:55.3266667+00:00
![](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(310.4, 87%, 40%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EDG%3C/text%3E%3C/svg%3E)
Dr. Golden
0
Reputation points
commented 2024-12-24T03:21:59.6433333+00:00
![](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(230.39999999999998, 61%, 32%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EPK%3C/text%3E%3C/svg%3E)
Prodip K. Saha
0
Reputation points
0 answers
Azure Speech JS SDK Returns Single Item in NBest Array
When using the Cognitive Services JavaScript Speech SDK with OutputFormat.Detailed and the recognizeOnceAsync approach, the NBest array consistently contains only a single object instead of the expected multiple alternatives. For example, when…
![](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(291.2, 48%, 38%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ETD%3C/text%3E%3C/svg%3E)
1 answer
Azure Speech - Custom Keyword
I'm trying to generate .table file but I'm not able to get a successfull training. Training model always give a "fail" state. I've even tried the "Hello Computer" keyword... it doesn't work. Am I doing something wrong ? =>…
asked 2021-04-29T10:02:40.157+00:00
Guillaume Demicheli
126
Reputation points
commented 2024-12-23T01:57:46.7766667+00:00
![](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(204.8, 23%, 29%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EMD%3C/text%3E%3C/svg%3E)
Michael Daubenmier
0
Reputation points
3 answers
How to use whisper model to transcribe audio in real time using Speech SDK?
How do I use Whisper model to transcribe microphone input in real-time using Microsoft-cognitiveservices-speech-sdk npm package? I currently have this working and my region is set to northcentralus I want to know how to use Whisper to transcribe in…
asked 2024-03-30T07:34:45.81+00:00
![](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(115.19999999999999, 69%, 21%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EN%3C/text%3E%3C/svg%3E)
Nas
0
Reputation points
answered 2024-12-21T05:49:32.9633333+00:00
![](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(198.4, 98%, 29%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EMR%3C/text%3E%3C/svg%3E)
Meezaan Ryklief👍🙂
0
Reputation points
1 answer
One of the answers was accepted by the question author.
Why i get wrong visemes when using German with English phrases?
I am using "Azure Speech" to synthesize speech from a text input, and also to generate Viseme. When using German language, if i use English phrase it sends me back wrong visemes. Ts is not good, last viseme has ts: 0, which should not happen.…
asked 2024-12-18T11:52:12.4866667+00:00
![](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(249.60000000000002, 27%, 34%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EVM%3C/text%3E%3C/svg%3E)
Veljko Markovic | Babylon Engineer
20
Reputation points
edited a comment 2024-12-19T20:07:13.07+00:00
![](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(246.4, 4%, 33%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ESS%3C/text%3E%3C/svg%3E)
Sina Salam
14,626
Reputation points
2 answers
How to access to Whisper in Azure Speech Batch Transcription
When listing base models from API I do not have the whisper option. How do I enable it? I am following this tutorial https://zcusa.951200.xyz/en-us/azure/ai-services/speech-service/batch-transcription-create?pivots=rest-api#use-a-whisper-model This is…
asked 2024-12-17T13:05:04.4933333+00:00
![](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(41.6, 20%, 14%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EVC%3C/text%3E%3C/svg%3E)
Václav Bílek
0
Reputation points
answered 2024-12-18T10:38:54.08+00:00
romungi-MSFT
48,221
Reputation points Microsoft Employee
1 answer
One of the answers was accepted by the question author.
"Internal server error" from Azure on Filipino speech-to-text
We have been using Azure speech-to-text / transcription services for generating transcripts. Recently (I first noticed this on Monday 9 December 2024) we have seen a very high chance of "Internal server error" in the transcription result of…
![](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(201.6, 26%, 29%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EJH%3C/text%3E%3C/svg%3E)
1 answer
One of the answers was accepted by the question author.
About speaker separation in "fast-transcription-api"
Dear Azure Support Team https://zcusa.951200.xyz/en-us/rest/api/speechtotext/transcriptions/transcribe?view=rest-speechtotext-2024-05-15-preview&tabs=HTTP The details of the TranscribeDefinition class are not described anywhere, so how should I do…
asked 2024-07-29T03:39:03.9533333+00:00
![](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(54.400000000000006, 90%, 15%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EYA%3C/text%3E%3C/svg%3E)
y.ashibe
45
Reputation points
commented 2024-12-18T09:08:29.2066667+00:00
![](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(156.8, 0%, 25%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ESB%3C/text%3E%3C/svg%3E)
Schuster, Björn
0
Reputation points
1 answer
One of the answers was accepted by the question author.
Fine-tuning speech-to-text base model for better address recognition
Hello, my team is creating a solution to transcribe addresses with higher accuracy. Our initial benchmarks for using a STT base model for address transcription suggests that it needs to be improved in order to be utilized in a production environment. I…
asked 2024-12-12T14:03:18.8133333+00:00
![](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(121.6, 69%, 21%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3ECC%3C/text%3E%3C/svg%3E)
Caesar Cavales
50
Reputation points
accepted 2024-12-17T13:21:38.8133333+00:00
Caesar Cavales
50
Reputation points
1 answer
One of the answers was accepted by the question author.
Handling Special Characters in Azure TTS Input
Hello, I’ve noticed that when sending text with special characters like \n (newline) to the Azure Text-to-Speech (TTS) engine, the output is synthesized literally as "backslash n." For now, we’re removing \n before sending the text to the TTS…
asked 2024-12-16T15:13:10.15+00:00
![](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(6.4, 26%, 10%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EAH%3C/text%3E%3C/svg%3E)
Ananth Hegde (anahegde)
20
Reputation points
accepted 2024-12-17T12:20:29.9566667+00:00
Ananth Hegde (anahegde)
20
Reputation points
1 answer
Data upload from Azure Blob Storage for the usage in Speech Studio
Hello everyone, In Speech Studio we can browse for the files locally. Is there a way to upload files from the Blob Storage? Might this functionality be present in Azure AI Foundry in the future? Many thanks in advance for your assistance!
asked 2024-12-11T11:08:25.4666667+00:00
![](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(185.6, 88%, 27%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EML%3C/text%3E%3C/svg%3E)
Mariana Logvinenko
20
Reputation points
commented 2024-12-16T09:14:12.2+00:00
![](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(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)
santoshkc
11,535
Reputation points Microsoft Vendor
0 answers
Azure Speech - Custom Keyword
I'm trying to generate .table file but I'm not able to get a successfull training. Training model give a "fail" state after more than 48 hours of tranning. the process is: => Speech Studio => Custom KeyWord => New Project => Train…
asked 2024-12-14T01:27:43.49+00:00
![](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(284.8, 73%, 37%)' 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)
shahsen
0
Reputation points
commented 2024-12-16T08:11:55.0333333+00:00
romungi-MSFT
48,221
Reputation points Microsoft Employee