# Custom ASR

Custom ASR enables users to integrate and test their own Automatic Speech Recognition (ASR) APIs with Datasaur projects. To enable this feature in your workspace, contact us at **<support@datasaur.ai>**.

## Configuring Custom ASR

### **Step 1: Set Up the Sample API**

1. Open the provided CodeSandbox example API [here](https://codesandbox.io/p/sandbox/datasaur-sample-custom-api-yf8g6p).
   * To test this code, you can also create a live endpoint using [codesandbox.io](http://codesandbox.io/).
2. Save any changes to the codebase to automatically create a fork for further testing and modifications.
3. Copy the **base URL** of the API from the CodeSandbox preview panel (on the right-hand side). In addition, you may need to ensure the panel is expanded if it’s collapsed.

   <figure><img src="https://448889121-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MbjY0HseEqu7LtYAt4d%2Fuploads%2Fgit-blob-a8234a2fdde35c69d3f962550ca18e5cd4b6d38f%2FCustom%20ASR%20-%20CodeSandbox.png?alt=media" alt=""><figcaption></figcaption></figure>
4. In the sample API that we provided, you can use the following endpoints based on your needs:
   * <mark style="color:red;">`{baseUrl}/text-extraction/example-text`</mark> (returns plain text transcription).
   * <mark style="color:red;">`{baseUrl}/text-extraction/example-json`</mark> (returns Importable JSON transcription).

### **Step 2: Add Custom API in Project Creation Wizard**

1. Go to the **Active projects** page and click **“Create project".**

   <figure><img src="https://448889121-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MbjY0HseEqu7LtYAt4d%2Fuploads%2Fgit-blob-86c29dd3ce7c46b99fb83cd9a616b08010b76315%2FProjects%20-%20active%20projects.png?alt=media" alt=""><figcaption></figcaption></figure>
2. In **Step 1: Upload**, upload your image/PDF (for OCR) or audio file (for ASR) and click **Next.**
3. In **Step 2: Preview**, click the dropdown labeled **“Apply ASR Method”** (or **“Apply OCR Method”** for images/PDFs).

   <figure><img src="https://448889121-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MbjY0HseEqu7LtYAt4d%2Fuploads%2Fgit-blob-7df43b9fe093be7d7e40f477848c3ce39ea4062d%2FPCW%20-%20Step%202%20-%20audio%20labeling%20-%20audio%20preview%20-%20apply%20ASR%20method.png?alt=media" alt=""><figcaption></figcaption></figure>
4. Select **“+ Add New API…”** to open a dialog box.

   <figure><img src="https://448889121-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MbjY0HseEqu7LtYAt4d%2Fuploads%2Fgit-blob-4577d0eb117b849f457e695d7ff4d23a317d8cbf%2FPCW%20-%20Step%202%20-%20audio%20labeling%20-%20audio%20preview%20-%20apply%20ASR%20method%20-%20dropdown.png?alt=media" alt=""><figcaption></figcaption></figure>

   * Fill in the fields:

     * **Name:** Desired name for the API.
     * **Custom API URL:** Paste one of the two endpoints from our sample API.
     * **Secret:** Add any placeholder value (this can be ignored for now).

     <figure><img src="https://448889121-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MbjY0HseEqu7LtYAt4d%2Fuploads%2Fgit-blob-cd6b8509c4479e8179641c35515cb8157b3b2a90%2FPCW%20-%20Step%202%20-%20audio%20labeling%20-%20add%20custom%20ASR%20API%20dialog.png?alt=media" alt=""><figcaption></figcaption></figure>
5. Click **Save** to add and select the API.
6. Review the sample transcription on the right panel.

   <figure><img src="https://448889121-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MbjY0HseEqu7LtYAt4d%2Fuploads%2Fgit-blob-2760c049c5551aafa44b7e0d19d8f566bceee250%2FPCW%20-%20Step%202%20-%20audio%20labeling%20-%20audio%20preview%20-%20apply%20ASR%20method%20changed.png?alt=media" alt=""><figcaption></figcaption></figure>

### **Step 3: Create the Project**

Once the API is added, complete the remaining project configuration steps to create the project and use the transcription results.

<figure><img src="https://448889121-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MbjY0HseEqu7LtYAt4d%2Fuploads%2Fgit-blob-4ded718a9b324db5a70b7a985c1fe1028c55aa38%2FProject%20-%20audio%20labeling%20-%20labeled.png?alt=media" alt=""><figcaption></figcaption></figure>

## ASR Sample File

{% file src="<https://448889121-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MbjY0HseEqu7LtYAt4d%2Fuploads%2Fgit-blob-84338de20b4630578209d11b8e58c2bf3df18939%2FDatasaur%20sample%20-%20ASR.mp3?alt=media&token=f9117b46-c6ef-4fa8-b765-92ac3f746d09>" %}

{% file src="<https://448889121-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2F-MbjY0HseEqu7LtYAt4d%2Fuploads%2Fgit-blob-a2266ed4e577b811339eeb855b10f883c6bc9966%2FDatasaur%20sample%20-%20Custom%20ASR%20(result).json?alt=media&token=472c0be9-a614-43b4-9547-82f300da9d8a>" %}

## Whisper ASR Limitations

<table data-full-width="false"><thead><tr><th width="263">Factor</th><th>Description</th></tr></thead><tbody><tr><td>Accepted file formats</td><td>.wav, .mp3, .m4a</td></tr><tr><td>Maximum individual file size</td><td>25 MB</td></tr><tr><td>Request limit (Azure)</td><td>100 RPM</td></tr><tr><td>Request timeout</td><td>10 minutes</td></tr><tr><td>Maximum API call retries</td><td>2 retries on request errors (e.g., rate limit reached, timeout).</td></tr></tbody></table>

## Amazon Transcribe ASR Limitations

<table><thead><tr><th width="272.546875">Factor</th><th>Description</th></tr></thead><tbody><tr><td>Accepted file formats</td><td>.flac, .m4a, .mp3, .wav</td></tr><tr><td>Maximum individual file size</td><td>130 MB</td></tr><tr><td>Maximum audio duration</td><td>14,400 seconds</td></tr></tbody></table>


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.datasaur.ai/api/custom-asr.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.