NOVA Voice Agent Starter Kit
Everything an AI agent needs to work by voice, from Nova, an open-source voice assistant built on Claude Code and Codex. The three skills cover the three jobs of a voice agent: hearing, speaking and acting safely.
nova-hinglish-stt sets up offline speech-to-text for English, Hindi and Hinglish, with Hindi written in Roman letters.
voice-replies makes the agent answer in one short spoken line, with the details kept on screen.
voice-agent-guard is a rulebook for a permission hook that decides which commands run, which ask first and which never run.
For developers building a voice assistant or a hands-free agent on Claude Code, Codex or a similar tool. On its own, each skill solves one piece. Together they cover the full loop: the user speaks, the agent acts within safe limits, and it answers in a way that sounds right out loud. The listening, speaking and microphone parts of your app are still yours to build.
Works with any agent that signs in to Agensi, such as Claude, ChatGPT, Claude Code, Cursor and Codex CLI.
One payment, lifetime access. 3 skills unlock instantly in your library.
What's included
3 skillsSpeech-to-text for English and Hinglish that outputs Romanized Hindi instead of Devanagari using faster-whisper.
A deterministic safety framework for AI agents to block, confirm, or allow shell commands and tool calls.
Format agent replies for natural-sounding text-to-speech by separating spoken summaries from on-screen code and data.
See it in action
NOVA Hinglish STT
Example session with this skill installed
I need a Python script to transcribe 'audio_input.wav' which contains a mix of Hindi and English. It must output the Hindi parts in Roman script.
- Read your context and instructions
- Compiled the nova hinglish stt
I have configured the transcription script using the Nova Whisper Hinglish model. This setup includes Voice Activity Detection (VAD) to handle silence and uses float16 compute for GPU efficiency.
Hinglish Transcription Script
import os
import sys
from faster_whisper import WhisperModel
# Ensure CUDA DLLs are in path for Windows GPU usage
# os.environ["PATH"] = r".venv\Lib\site-packages\nvidia\cublas\bin;" + os.environ["PATH"]
def transcribe_hinglish(audio_path):
# Initialize model on GPU with float16 for speed
model = WhisperModel(
"AyushM1701/nova-whisper-hinglish",
device="cuda",
compute_type="float16"
)
# Custom VAD to avoid clipping soft word endings
vad_params = {
"threshold": 0.35,
"speech_pad_ms": 300,
"min_silence_duration_ms": 500
}
# language="hi" forces Romanized Hindi output
segments, info = model.transcribe(
audio_path,
language="hi",
vad_filter=True,
vad_parameters=vad_params,
beam_size=1,
condition_on_previous_text=False
)
return " ".join(s.text.strip() for s in segments)
if __name__ == "__main__":
result = transcribe_hinglish("audio_input.wav")
print(f"Transcription: {result}")
Next steps
- Install dependencies using
pip install faster-whisper "av<16". - If running on Windows with a GPU, verify the CUDA 12 bin folders are added to your environment
PATH. - Run the script; the first execution will download the 1.6 GB model weights.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads these skills and uses them on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every skill in this bundle goes through our trust checks.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.