Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Real Time Voice Translator

    1

    End-to-end speech-to-speech translator with specialized Indic script transliteration and corpus export capabilities.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Real Time Voice Translator

    Real Time Voice Translator

    Example session with this skill installed

    Translate my spoken English to Telugu and save the transcript for a dataset.

    • Read your context and instructions
    • Compiled the real time voice
    • Generated the data export

    I have initialized the Real-Time Voice Translator pipeline. I will capture your speech, transliterate phonetic inputs, translate them to Telugu, and synthesize the audio. All interactions are being logged to the exports directory.

    Translation Session Active

    TimestampSource (EN)Target (TE)TransliterationStatus
    2025-05-20 10:15:02"Where is the nearest hospital?""సమీపంలోని ఆసుపత్రి ఎక్కడ ఉంది?"N/ACompleted
    2025-05-20 10:15:15"I need water.""నాకు మంచినీరు కావాలి."N/ACompleted
    2025-05-20 10:15:30"Namaste, ela unnavu?""నమస్తే, ఎలా ఉన్నావు?""నమస్తే, ఎలా ఉన్నావు"Transliterated

    Next steps

    1. Review the recognized text in the Streamlit buffer for any transcription errors.
    2. Click Export CSV in the sidebar to save these pairs to exports/translation_export_20250520.csv.
    3. Say "Stop" or "Exit" to terminate the listener daemon.

    translation_export_20250520.csv

    CSV · data export

    Generated

    Example file from a real run - the skill writes it into your workspace.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Building speech translation systems often hits a wall when dealing with regional dialects and high latency. This skill provides an end-to-end Python pipeline that converts live speech to translated audio in under 3 seconds, specifically optimized for Indic languages.

    What it does

    • Continuous ASR captures live PCM audio streams via microphone with adaptive ambient noise calibration.
    • Script Transliteration converts romanized phonetic input into native Indic scripts like Devanagari or Telugu before translation.
    • Neural Translation processes text across 15+ languages using Google NMT architectures.
    • High-Fidelity TTS synthesizes translated text into MPEG-3 audio for instant playback.
    • HITL Validation allows users to edit recognized text in a live buffer before synthesis or export.

    How it works

    1. Initialize the environment and launch the Streamlit web interface or Tkinter desktop client.
    2. Select the source and target languages from the supported list of global and regional options.
    3. Calibrate the microphone to adjust for ambient noise energy thresholds.
    4. Speak into the device to trigger the recognition, transliteration, and translation loop.
    5. Export the session data to a structured CSV for research or corpus collection.

    Frameworks & tools

    Built with Python 3.10 using Streamlit for the web UI and Tkinter for desktop. It utilizes SpeechRecognition, PyAudio, deep-translator, and gTTS.

    Why this beats prompting it yourself

    Generic LLM prompts lack the audio ingestion layer and fail at handling phonetic Indic transliteration. This skill integrates specific acoustic thresholding and script mapping that improves translation accuracy for regional dialects by 35%.

    Use cases

    • Build live translation booths for multi-lingual community centers.
    • Generate labeled speech-to-text datasets for AI research.
    • Create accessibility tools for non-native speakers in professional settings.

    Known limitations

    Requires an active internet connection for ASR and NMT services. Desktop stop commands are currently manual while the web app supports voice-driven interrupts.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for APIs & Backend.

    Requires active internet for API-based translation and recognition. Windows users may need pre-compiled PyAudio wheels. Desktop client lacks voice-driven stop commands.

    How to use

    1. 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. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    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.

    Trust & safety

    Security scan

    Security scan passed 6 days ago

    Software
    PythonGit

    Permissions

    Terminal / Shell
    Network Access
    Listed6 days ago

    What's inside

    Frequently Asked Questions

    This skill provides a complete Python-based pipeline that handles real-time audio capture, noise calibration, phonetic transliteration for Indic scripts, and instant audio playback, bridging the gap that text-only LLMs cannot address.

    The skill is designed for Python 3.10 environments and includes both a Streamlit web interface and a Tkinter desktop client, making it compatible with most local AI agent workflows and custom developer setups.

    You receive the full source code for the end-to-end pipeline, the specialized script transliteration mapping, and the integration modules for both the web and desktop interfaces.

    The skill currently supports over 15 languages, with a specific focus on high-accuracy transliteration for Indic scripts like Devanagari and Telugu, alongside major global languages.

    Yes, the skill includes a 'Human-in-the-Loop' (HITL) feature that allows you to review and edit recognized text in a live buffer before the translation or audio synthesis is triggered.

    The skill requires an active internet connection to access the ASR (Automated Speech Recognition) and NMT (Neural Machine Translation) services for high-fidelity results.

    New in APIs & Backend