Skip to content
TakoAPI
Back to skills

miranda-elevenlabs-speech

by jeffpignataro

Communication
—Not yet rated
Your rating

Brief

Text-to-Speech and Speech-to-Text using ElevenLabs AI.

Description

What This Skill Does

Provides both text-to-speech and speech-to-text using the ElevenLabs API. TTS supports multiple AI voices with adjustable stability and similarity settings. STT uses ElevenLabs Scribe and handles 99 languages, speaker diarization, and common audio formats.

Combining TTS and STT in one API client eliminates the need to integrate separate services for voice input and output.

When to Use It

  • Convert a written article to an audio file for listening on the go
  • Transcribe a recorded meeting or voice note into text
  • Reply to a Telegram voice message with a synthesized voice response
  • Generate multilingual audio content using the eleven_multilingual_v2 model
  • Detect and separate speakers in a multi-person audio recording

Example Workflow

INPUT

User asks: transcribe this voice message and reply with audio

AGENT
  1. 1

    Receive the audio file (e.g. user_voice.ogg)

  2. 2

    Transcribe it using ElevenLabsScribe.transcribe()

  3. 3

    Apply any processing logic to generate a response text

  4. 4

    Convert the response text to speech using ElevenLabsClient.text_to_speech()

  5. 5

    Send the resulting MP3 as a voice reply

OUTPUT

An MP3 audio file containing the spoken response, sent back to the user

Requirements

  • ELEVENLABS_API_KEY environment variable
  • ElevenLabs account (free tier available at elevenlabs.io)

Installation

clawhub install jeffpignataro/miranda-elevenlabs-speech

Run the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.

Links