OpenIntegrations/docs/en/md/OpenAI/Audio-processing/Generate-speech.mdx

---
sidebar_position: 1
description: Generate speech and other functions to work with OpenAI in the Open Integration Package, a free open-source integration library for 1C:Enterprise 8, OneScript and CLI
keywords: [1C, 1С, 1С:Enterprise, 1С:Enterprise 8.3, API, Integration, Services, Exchange, OneScript, CLI, OpenAI]
---

import Tabs from '@theme/Tabs';
import TabItem from '@theme/TabItem';

# Generate speech
 Generates audio with the specified text for speech synthesis



`Function GenerateSpeech(Val URL, Val Token, Val Model, Val Text, Val Voice = "alloy", Val AdditionalParameters = "", Val AdditionalHeaders = "") Export`

  | Parameter | CLI option | Type | Required | Description |
  |-|-|-|-|-|
  | URL | --url | String | ✔ | OpenAI server URL |
  | Token | --token | String | ✔ | OpenAI authorization token |
  | Model | --model | String | ✔ | Models name |
  | Text | --input | String | ✔ | Text for speech synthesis |
  | Voice | --voice | String | ✖ | Voice type: alloy, ash, ballad, coral, echo, etc.. |
  | AdditionalParameters | --options | Structure Of KeyAndValue | ✖ | Additional request parameters, if necessary |
  | AdditionalHeaders | --headers | Map Of KeyAndValue | ✖ | Additional request headers, if necessary |

  
  Returns:  BinaryData - Processing result

<br/>

:::tip
Method at API documentation: [Create speech](https://platform.openai.com/docs/api-reference/audio/createSpeech)

Available voices may vary depending on the selected model

The audio file format of the response can be changed by adding `response_format` in additional parameters. Available formats: mp3 (default), opus, aac, flac, wav, pcm
:::
<br/>



```bsl title="1C:Enterprise/OneScript code example"
    URL   = "https://hut.openintegrations.dev/localai/";
    Token = "12We...";

    Text  = "Attack ships on fire off the shoulder of Orion bright as magnesium";
    Model = "tts-1";

    AdditionalParameters = New Structure("response_format", "wav");

    Result = OPI_OpenAI.GenerateSpeech(URL, Token, Model, Text, , AdditionalParameters);
```
 

 <Tabs>
  
    <TabItem value="bash" label="Bash" default>
        ```bash
            # JSON data can also be passed as a path to a .json file
            
            oint openai GenerateSpeech \
             --url "https://hut.openintegrations.dev/localai/" \
             --token "***" \
             --model "tts-1" \
             --input "Attack ships on fire off the shoulder of Orion bright as magnesium" \
             --options "{'response_format':'wav'}"
        ```
    </TabItem>
  
    <TabItem value="bat" label="CMD/Bat" default>
        ```batch
            :: JSON data can also be passed as a path to a .json file
            
            oint openai GenerateSpeech ^
             --url "https://hut.openintegrations.dev/localai/" ^
             --token "***" ^
             --model "tts-1" ^
             --input "Attack ships on fire off the shoulder of Orion bright as magnesium" ^
             --options "{'response_format':'wav'}"
        ```
    </TabItem>
</Tabs>


```json title="Result"
NOT JSON: 52 49 46 46 24 53 02 00 57 41 56 45 66 6D 74 20 10 00 00 00 01 00 01 00 80 3E 00 00 00 7D 00 00 02 00 10 00 64 61 74 61 00 53 02 00 FE FF 05 00 04 00 00 00 00 00 04 00 07 00 04 00 01 00 00 00…
```