Skip to content
TakoAPI
Back to skills

zvukogram

by erview

Search & Research
471Downloads reported by clawskills.sh
2Stars
0Views
0 likesSign in to like this skill
—Not yet rated
Your rating

Brief

Text-to-Speech via Zvukogram API with SSML support.

Description

What This Skill Does

Generates speech from text using the Zvukogram API. Supports SSML markup for stress marks, pauses, and speed control. Can merge multiple audio fragments and handle English word transcription for Russian TTS.

SSML support with per-word stress control and English transcription aliases gives more accurate Russian-language output than basic TTS APIs.

When to Use It

  • Converting written articles to audio files
  • Adding voice notifications to automated pipelines
  • Recording podcast narration with multiple voices
  • Voicing news scripts with proper pronunciation control
  • Generating audio from long-form text documents

Example Workflow

INPUT

User asks: 'Read this news summary aloud with a pause after the headline, using a female voice'

AGENT
  1. 1

    Wraps headline in SSML with a <break time='500ms'/> after it

  2. 2

    Adds <sub alias='...'/> tags for any English brand names in the text

  3. 3

    Calls tts.py with --voice Алена and the SSML-formatted text

  4. 4

    Saves output to a local .mp3 file

OUTPUT

An .mp3 file with the news summary spoken by Алена, pausing after the headline and pronouncing brand names correctly

Requirements

  • ZVUKOGRAM_TOKEN environment variable (obtain at https://zvukogram.com/)
  • ZVUKOGRAM_EMAIL environment variable (your Zvukogram account email)

Installation

clawhub install erview/zvukogram

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

Links