how to call azure service to generate text to voice

how to call api https://centralindia.tts.speech.microsoft.com/cognitiveservices/v1 using postman

ChatGPT

To call the API at https://centralindia.tts.speech.microsoft.com/cognitiveservices/v1 using Postman, follow these steps:

Prerequisites

1. Subscription Key: You need a valid subscription key from Azure Cognitive Services.
2. Authentication: Ensure you have the necessary permissions and have your subscription key ready.

Steps to Make a POST Request

1. Open Postman:
• Launch Postman on your computer.
2. Create a New Request:
• Click on New and select Request.
3. Set the Request Type to POST:
• In the new request window, set the HTTP method to POST.
4. Enter the Request URL:
• Enter https://centralindia.tts.speech.microsoft.com/cognitiveservices/v1 in the URL field.
5. Set Up Headers:
• Click on the Headers tab and add the following headers:
• Ocp-Apim-Subscription-Key: Your subscription key.
• Content-Type: application/ssml+xml
• X-Microsoft-OutputFormat: riff-24khz-16bit-mono-pcm (or any other output format you need).
6. Create the Request Body:
• Click on the Body tab.
• Select the raw option.
• Ensure the type is XML.
• Enter your SSML (Speech Synthesis Markup Language) in the body. Here is an example of what your SSML might look like:

xml

Copy code

<speak version='1.0' xml:lang='en-US'>

    <voice xml:lang='en-US' xml:gender='Female' name='en-US-AriaNeural'>

        Hello, this is a test.

    </voice>

</speak>

7. Send the Request:
• Click the Send button to make the request.

Example Postman Setup

Headers

Key	Value
Ocp-Apim-Subscription-Key	your_subscription_key
Content-Type	application/ssml+xml
X-Microsoft-OutputFormat	riff-24khz-16bit-mono-pcm

Body

xml

Copy code

<speak version='1.0' xml:lang='en-US'>

    <voice xml:lang='en-US' xml:gender='Female' name='en-US-AriaNeural'>

        Hello, this is a test.

    </voice>

</speak>

Expected Response

• If successful, the response will be the audio data in the specified format.

Troubleshooting

• 401 Unauthorized: Check your subscription key.
• 400 Bad Request: Check the format of your SSML and headers.
• 500 Internal Server Error: Try again later or check the Azure status.

By following these steps, you should be able to successfully call the Text-to-Speech API using Postman.

4o

how to select speed for above?

ChatGPT

To control the speaking speed in the SSML for the Text-to-Speech API, you can use the <prosody> element within your SSML. The rate attribute of the <prosody> element allows you to adjust the speaking speed.

Here's how you can modify your SSML to include the rate attribute to control the speed:

Example SSML with Speed Adjustment

xml

Copy code

<speak version='1.0' xml:lang='en-US'>

    <voice xml:lang='en-US' xml:gender='Female' name='en-US-AriaNeural'>

        <prosody rate="0.75">Hello, this is a test spoken slowly.</prosody>

    </voice>

</speak>

In this example, rate="0.75" means 75% of the default speed (i.e., slower). You can adjust the rate attribute to different values to control the speed:

• rate="1.0" is the default speed.
• rate="0.75" is 75% of the default speed (slower).
• rate="1.25" is 125% of the default speed (faster).
• You can also use relative values like rate="slow", rate="medium", rate="fast".

Updated Postman Setup

Headers

Key	Value
Ocp-Apim-Subscription-Key	your_subscription_key
Content-Type	application/ssml+xml
X-Microsoft-OutputFormat	riff-24khz-16bit-mono-pcm

Body

xml

Copy code

<speak version='1.0' xml:lang='en-US'>

    <voice xml:lang='en-US' xml:gender='Female' name='en-US-AriaNeural'>

        <prosody rate="0.75">Hello, this is a test spoken slowly.</prosody>

    </voice>

</speak>

Steps in Postman

1. Open Postman.
2. Create a new request and set the HTTP method to POST.
3. Enter the request URL: https://centralindia.tts.speech.microsoft.com/cognitiveservices/v1.
4. Set the headers:
• Ocp-Apim-Subscription-Key: Your subscription key.
• Content-Type: application/ssml+xml
• X-Microsoft-OutputFormat: riff-24khz-16bit-mono-pcm
5. Set the body to raw and XML:

xml

Copy code

<speak version='1.0' xml:lang='en-US'>

    <voice xml:lang='en-US' xml:gender='Female' name='en-US-AriaNeural'>

        <prosody rate="0.75">Hello, this is a test spoken slowly.</prosody>

    </voice>

</speak>

6. Send the request.

This configuration will produce a speech output with the adjusted speed based on the rate value you have set in the SSML.

4o

how to get free subscription key from microsfot for above?

ChatGPT

To get a free subscription key for Microsoft's Text-to-Speech (TTS) API, you need to sign up for an Azure account and create a Cognitive Services resource. Microsoft offers a free tier that includes some free usage of the TTS API. Here are the steps to obtain your free subscription key:

Steps to Get a Free Subscription Key

1. Sign Up for an Azure Account:
• Go to the Azure website.
• Click on the "Start free" button to create a new Azure account.
• Follow the on-screen instructions to set up your account. You'll need to provide some basic information and a credit card for verification (no charges will be made for the free tier).
2. Create a Cognitive Services Resource:
• Once your Azure account is set up, log in to the Azure Portal.
• In the Azure Portal, click on the "Create a resource" button.
• In the search bar, type "Cognitive Services" and select it from the list.
• Click on the "Create" button.
3. Configure the Cognitive Services Resource:
• Subscription: Select your Azure subscription.
• Resource Group: Create a new resource group or select an existing one.
• Region: Choose the region closest to you (ensure it's the same as where you intend to use the service, e.g., Central India).
• Name: Give your Cognitive Services resource a unique name.
• Pricing Tier: Select the free tier (F0) if available. This tier provides some free usage each month.
• Review + Create: Review your settings and click on the "Create" button.
4. Get Your Subscription Key:
• After the resource is created, navigate to it in the Azure Portal.
• Under the "Keys and Endpoint" section, you will find two keys. You can use either key for your API requests.
• Copy one of the subscription keys.

Using the Subscription Key in Postman

Now that you have your subscription key, you can use it in Postman as described in the previous steps:

1. Open Postman.
2. Create a new request and set the HTTP method to POST.
3. Enter the request URL: https://centralindia.tts.speech.microsoft.com/cognitiveservices/v1.
4. Set the headers:
• Ocp-Apim-Subscription-Key: Your subscription key (copied from Azure).
• Content-Type: application/ssml+xml
• X-Microsoft-OutputFormat: riff-24khz-16bit-mono-pcm
5. Set the body to raw and XML:

xml

Copy code

<speak version='1.0' xml:lang='en-US'>

    <voice xml:lang='en-US' xml:gender='Female' name='en-US-AriaNeural'>

        <prosody rate="0.75">Hello, this is a test spoken slowly.</prosody>

    </voice>

</speak>

6. Send the request.

By following these steps, you can obtain a free subscription key and use it to make requests to the Text-to-Speech API with Postman.

4o

output as mp3 ? for the above request postman reqeust?

ChatGPT

To receive the output as an MP3 file from the Microsoft Text-to-Speech API, you need to modify the X-Microsoft-OutputFormat header to specify an MP3 output format. Here are the steps to achieve this using Postman:

Steps to Make a POST Request for MP3 Output

1. Open Postman:
• Launch Postman on your computer.
2. Create a New Request:
• Click on New and select Request.
3. Set the Request Type to POST:
• In the new request window, set the HTTP method to POST.
4. Enter the Request URL:
• Enter https://centralindia.tts.speech.microsoft.com/cognitiveservices/v1 in the URL field.
5. Set Up Headers:
• Click on the Headers tab and add the following headers:
• Ocp-Apim-Subscription-Key: Your subscription key.
• Content-Type: application/ssml+xml
• X-Microsoft-OutputFormat: audio-16khz-128kbitrate-mono-mp3
6. Create the Request Body:
• Click on the Body tab.
• Select the raw option.
• Ensure the type is XML.
• Enter your SSML (Speech Synthesis Markup Language) in the body. Here is an example of what your SSML might look like:

xml

Copy code

<speak version='1.0' xml:lang='en-US'>

    <voice xml:lang='en-US' xml:gender='Female' name='en-US-AriaNeural'>

        <prosody rate="0.75">Hello, this is a test spoken slowly.</prosody>

    </voice>

</speak>

7. Send the Request:
• Click the Send button to make the request.

Example Postman Setup

Headers

Key	Value
Ocp-Apim-Subscription-Key	your_subscription_key
Content-Type	application/ssml+xml
X-Microsoft-OutputFormat	audio-16khz-128kbitrate-mono-mp3

Body

xml

Copy code

<speak version='1.0' xml:lang='en-US'>

    <voice xml:lang='en-US' xml:gender='Female' name='en-US-AriaNeural'>

        <prosody rate="0.75">Hello, this is a test spoken slowly.</prosody>

    </voice>

</speak>

Expected Response

• If successful, the response will be the audio data in MP3 format. You can save this data to an MP3 file.

Saving the Response as an MP3 File

1. Send the Request: Click the Send button in Postman.
2. Save the Response:
• After receiving the response, click on the Save Response button (an icon that looks like a download button) in Postman.
• Choose a location to save the file and save it with an .mp3 extension.

By following these steps, you should be able to receive and save the Text-to-Speech output as an MP3 file using Postman.