Feat : Venice audio (Text to Speech)

[yornfifty]

Description

Please include a summary of the changes and the related issue.

Text to speech with venice ai, support multiple voice models

af_alloy, af_aoede, af_bella, af_heart, af_jadzia, af_jessica, af_kore, af_nicole, af_nova, af_river, af_sarah, af_sky, am_adam, am_echo, am_eric, am_fenrir, am_liam, am_michael, am_onyx, am_puck, am_santa, bf_alice, bf_emma, bf_lily, bm_daniel, bm_fable, bm_george, bm_lewis, zf_xiaobei, zf_xiaoni, zf_xiaoxiao, zf_xiaoyi, zm_yunjian, zm_yunxi, zm_yunxia, zm_yunyang, ff_siwis, hf_alpha, hf_beta, hm_omega, hm_psi, if_sara, im_nicola, jf_alpha, jf_gongitsune, jf_nezumi, jf_tebukuro, jm_kumo, pf_dora, pm_alex, pm_santa, ef_dora, em_alex, em_santa

Utils.s3 update

now we can use store_file_bytes to store most common file type and ability to limit the filesize

class FileType(str, Enum):
    IMAGE = "image"
    VIDEO = "video"
    AUDIO = "audio"
    PDF = "pdf"


async def store_file_bytes(
    file_bytes: bytes,
    key: str,
    file_type: FileType,
    size_limit_bytes: Optional[int] = None,
) -> str:

Type of Change

• Bugfix
• New Feature
• Improvement
• Documentation Update

Checklist

• I have read the contributing guidelines.
• I have added tests to cover my changes.
• All new and existing tests passed.

*Note

i'm not add this to models/agent_schema.json's property yet, to avoid conflic with other pr

tell me if i should add this right away

        "venice_audio": {
          "title" : "Venice Audio",
          "$ref": "../skills/venice_audio/schema.json"
        },

Related Issue

• Skills : Venice Audio (Text to Speech) #503
• Skill: Generate voice from text input #314

Showcase

i made a simple chat app to interact with the agent to showcase the functionality
https://crestal.s3.ap-southeast-1.amazonaws.com/local/intentkit/2025-04-27%2003-40-17.mp4

[bluntbrain]

Hi @yornfifty , great work here!

I see these 2 minor issues, this PR can be merged as soon as these are fixed, thanks!

1. Config class incorrectly uses TypedDict instead of inheriting from SkillConfig:

# Current problematic code:
class Config(TypedDict):
    enabled: bool
    api_key: str
    states: SkillStates

# Should be changed to:
class Config(SkillConfig):
    """Configuration for Venice Audio skills."""
    enabled: bool
    api_key: str
    states: SkillStates

2. Missing proper base.py implementation:

# Missing required base.py file with proper inheritance pattern:
from typing import Type
from pydantic import BaseModel, Field
from abstracts.skill import SkillStoreABC
from skills.base import IntentKitSkill

class VeniceAudioBaseTool(IntentKitSkill):
    """Base class for Venice Audio tools."""
    
    name: str = Field(description="The name of the tool")
    description: str = Field(description="A description of what the tool does")
    args_schema: Type[BaseModel]
    skill_store: SkillStoreABC = Field(description="The skill store for persisting data")
    
    @property
    def category(self) -> str:
        return "venice_audio"

[yornfifty]

thank you for the review, i already fix it, should i also add it to agent_schema.json?

[hyacinthus]

You can add a enum field to config, and get config in _arun in runtime...

Too many opinions here.

[yornfifty]

working on it

[yornfifty]

i just deep dive to how draft-07 schema works, and i've come to this
what do you think?

{
    "$schema": "http://json-schema.org/draft-07/schema#",
    "title": "Venice Audio Skills",
    "description": "Configuration for the Venice Audio skill.",
    "type": "object",
    "x-tags": [
        "AI",
        "Audio",
        "Text to Speech"
    ],
    "properties": {
        "enabled": {
            "type": "boolean",
            "description": "Enable or disable the Venice Audio skill.",
            "default": false
        },
        "voice_model": {
            "type": "string",
            "enum": [
                "af_heart",
                "bm_lewis",
                "custom"
            ],
            "x-enum-title": [
                "af_heart (default female)",
                "bm_lewis (default male)",
                "Custom"
            ],
            "description": "Text to speech tool",
            "default": "disabled"
        },
        "states": {
            "type": "object",
            "title": "Skill States",
            "description": "Enable/disable specific voice models. Only enable one if you want a consistent characteristic for your agent. See docs for voice details and quality grades.",
            "properties": {
                "text_to_speech": {
                    "type": "string",
                    "enum": [
                        "disabled",
                        "public",
                        "private"
                    ],
                    "x-enum-title": [
                        "Disabled",
                        "Agent Owner + All Users",
                        "Agent Owner Only"
                    ],
                    "description": "Text to speech tool",
                    "default": "disabled"
                }
            }
        },
        "api_key_provider": {
            "type": "string",
            "title": "API Key Provider",
            "description": "Provider of the API key",
            "enum": [
                "agent_owner"
            ],
            "x-enum-title": [
                "Owner Provided"
            ],
            "default": "agent_owner"
        }
    },
    "required": [
        "states",
        "enabled"
    ],
    "allOf": [
        {
            "if": {
                "allOf": [
                    {
                        "properties": {
                            "enabled": {
                                "const": true
                            }
                        }
                    },
                    {
                        "properties": {
                            "api_key_provider": {
                                "const": "agent_owner"
                            }
                        }
                    }
                ]
            },
            "then": {
                "properties": {
                    "api_key": {
                        "type": "string",
                        "title": "Venice API Key",
                        "x-link": "[Get your API key](https://venice.ai/)",
                        "x-sensitive": true,
                        "description": "API Key for authenticating with the Venice AI API."
                    }
                },
                "required": [
                    "api_key"
                ]
            }
        },
        {
            "if": {
                "properties": {
                    "voice_model": {
                        "const": "custom"
                    }
                }
            },
            "then": {
                "properties": {
                    "voice_model_custom": {
                        "type": "array",
                        "items": {
                            "type": "string"
                        },
                        "title": "Voice Model (Custom)",
                        "x-link": "[Supported Voice Model](https://docs.venice.ai/api-reference/endpoint/audio/speech#body-voice)",
                        "description": "You can add one or more custom voice models.",
                        "default": [
                            "af_heart", "bm_lewis"
                        ]
                    }
                },
                "required": [
                    "voice_model_custom"
                ]
            }
        }
    ],
    "additionalProperties": true
}

[yornfifty]

latest test screenshot