> ## Documentation Index
> Fetch the complete documentation index at: https://docs.klyra.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# AI Speech to Text

> Quickly and accurately convert your audio into text with our AI Speech to Text feature. Supports multiple file formats and 25+ languages for reliable transcription in just minutes.

## Introduction

The [**AI Speech to Text**](https://klyra.ai/page/ai-speech-to-text) feature allows you to seamlessly transcribe audio and video files into accurate, editable text. Whether you’re working with **mp3 lectures, wav interviews, m4a podcasts, or mp4 recordings**, Klyra AI supports a wide range of formats.

With **25+ supported languages**, automatic detection, and the ability to export transcriptions in multiple formats, this tool is ideal for **content creators, professionals, educators, and researchers** who need fast and reliable transcription.

## How to Use

> 🔎 **Navigation:** Dashboard → Left Panel → AI Audio & Voice → Speech to Text

<Steps>
  <Step title="Upload your audio file">
    * Drag & Drop or click **Browse** to upload your audio file.
    * **Supported formats:** mp3, mp4, mpeg, mpga, m4a, wav, webm.
  </Step>

  <Step title="Add Audio Description (Optional)">
    * Provide a short description of your audio file (e.g., *“Team meeting on project roadmap”*).
    * This helps AI contextualize and improve transcription accuracy.
  </Step>

  <Step title="Set Audio Language (Optional)">
    * By default, the tool uses **Auto Detect**.
    * If needed, you can select the language manually from the dropdown list.
  </Step>

  <Step title="(Optional)">
    * **Transcribe Audio File** (default): Transcribes audio into the original language.
    * **Generate Transcription in English**: Translates and transcribes your file into English.
  </Step>

  <Step title="Assign to a Workbook">
    Save your transcription in an organized folder under **AI Panel → Documents → All Transcripts**.
  </Step>

  <Step title="Name Your File">
    * By default, files are named **“Audio File”**.
    * You can give it a more descriptive title for easier tracking.
  </Step>

  <Step title="Start Transcription">
    * Click the **Transcribe** button.
    * The transcription process will begin, and progress will be shown in the workspace.
  </Step>

  <Step title="View & Edit Results">
    * Once transcription is complete, results will appear in the **right-hand workspace**.
    * You can directly **edit the text** to correct names, context, or special terms.

          <img src="https://mintcdn.com/klyraai/cTaTCqHYcMkGrUVF/images/klyra-ai-dashboard-speech-to-text.webp?fit=max&auto=format&n=cTaTCqHYcMkGrUVF&q=85&s=fb27e70b255dd16fccaf3c8845529173" alt="Klyra Ai Dashboard Speech To Text Web" class="border rounded-lg shadow-md" width="1920" height="1362" data-path="images/klyra-ai-dashboard-speech-to-text.webp" />
  </Step>

  <Step title="Save & Export">
    * After editing, click **Save**. The transcription will be stored under:\
      **AI Panel → Documents → All Transcripts**.
    * Export options include:
      * **Word Document (DOCX)**
      * **Text File (TXT)**
      * **Copy to Clipboard**
  </Step>
</Steps>

### 💡 Tips

* **Shorten background noise:** Clean audio improves accuracy.
* **Language hint:** If your file has mixed accents or languages, specify the main language instead of relying on Auto Detect.
* **Use Workbooks:** Organize transcriptions by projects (meetings, interviews, lectures).
* **Quick edit in-app:** Fix names, jargon, or technical terms right in the editor before exporting.
* **Save in DOCX for formatting:** Best for sharing with teams or editing in Word.

### 📌 Example Use Cases

* **Journalists:** Transcribe interviews into editable text.
* **Students:** Convert lectures and seminars into study notes.
* **Businesses:** Document team meetings, presentations, or training sessions.
* **Podcasters:** Generate transcripts for SEO and accessibility.
* **Researchers:** Transcribe focus groups, field interviews, or conference talks.