Azerbaijani Voiceover Studio

    1

    Brands are respelled (iPhone → Ayfon), emoji removed.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Azerbaijani Voiceover Studio

    Example session with this skill installed

    Kafemiz ucun reklam sesi lazimdir, kisi sesi, zarafatli olsun. Musteri ucun istifade edecem. Metn: "Seher yemeyi saat 8:00-dan 11:30-a qeder! Cay + 2 qutab cemi 4,50 AZN. Unvan: Xetai pr.
    15. Sifaris: 055 987 65 43"

    • Read your context and instructions
    • Compiled the azerbaijani voiceover studio

    Fictional café, address and phone number.

    Bu reklamdır (kommersiya), ona görə ElevenLabs-in pulsuz planı olmaz. Gemini və ya Azure istifadə et.

    Səs: Kişi · Stil: 10 Ceyhun (zarafatcıl vloq) · 35 söz ≈ 15 s

    Səhər yeməyi saat səkkizdən on bir otuza qədər! Çay və iki qutab, cəmi dörd manat əlli qəpik. Ünvan: Xətai prospekti, on beş. Sifariş: sıfır əlli beş, doqquz yüz səksən yeddi, altmış beş, qırx üç.
    

    Dəyişikliklər: Seher yemeyi → Səhər yeməyi · Cay → Çay · cemi → cəmi · Unvan → Ünvan · Xetai → Xətai · Sifaris → Sifariş · 8:00-dan → səkkizdən · 11:30-a → on bir otuza · + → və · 2 → iki · 4,50 AZN → dörd manat əlli qəpik · pr. 15 → prospekti, on beş · telefon qruplarla oxunur

    Gemini (AI Studio): 1) aistudio.google.com → Generate speech. 2) Səs:

    Puck (və ya Charon). Bloku yapışdır. 3) Run → dinlə → yüklə.

    Azerbaijani, native Baku accent, casual and colloquial. Playful, funny young male vlogger. Fast pace about 145 words per minute, lively ups and downs, but read the phone number slowly and clearly:
    Səhər yeməyi saat səkkizdən on bir otuza qədər! Çay və iki qutab, cəmi dörd manat əlli qəpik. Ünvan: Xətai prospekti, on beş. Sifariş: sıfır əlli beş, doqquz yüz səksən yeddi, altmış beş, qırx üç.
    

    Azure (commercial-safe, Babek): speech.microsoft.com → Audio Content Creation → New file → SSML view → paste → Play → Export.

    <speak version="1.0" xmlns="http://www.w3.org/2001/10/synthesis" xml:lang="az-AZ">
      <voice name="az-AZ-BabekNeural">
        <prosody rate="+15%" pitch="+5%">Səhər yeməyi saat səkkizdən on bir otuza qədər!</prosody>
        <break time="300ms"/>
        <prosody rate="+12%" pitch="+3%">Çay və iki qutab,<break time="150ms"/> cəmi dörd manat əlli qəpik.</prosody>
        <break time="300ms"/>
        <prosody rate="+5%" pitch="+0%">Ünvan: Xətai prospekti,<break time="150ms"/> on beş.</prosody>
        <break time="400ms"/>
        <prosody rate="-10%" pitch="+0%">Sifariş: sıfır əlli beş,<break time="200ms"/> doqquz yüz səksən yeddi,<break time="200ms"/> altmış beş,<break time="200ms"/> qırx üç.</prosody>
      </voice>
    </speak>
    

    Kommersiya: Gemini pulsuz versiyada qadağan etmir. Məlumatlar Google tərəfindən istifadə oluna bilər və səsdə SynthID nişanı olur. Azure F0 kommersiya üçün açıqdır, amma qeydiyyat üçün kart lazımdır. ElevenLabs Free kommersiya üçün olmaz, Starter planı (≈ $6/ay) lazımdır.
    AI qeydi: "Reklamdakı səs süni intellektlə yaradılıb."

    Dinləyəndə "ş" və "ğ" səslərinə fikir ver: Babek bəzən onları "s" və "q" kimi deyir. Belə olsa, Gemini-də Puck səsini sına.

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

    About this skill

    Write Azerbaijani text and get a voiceover script that free text-to-speech engines read naturally, with male and female voices, correct ç ş ə ğ ı ö ü, and no robotic number reading.

    Why it exists

    Most "robotic" Azerbaijani AI voiceovers are caused by the text, not the voice: digits and symbols ("24 saat", "$150", "30%", "18:30", phone numbers) are read in the wrong language, text typed without ə/ş/ç turns ə into e and ş into s, Cyrillic text and English brand names confuse the engine, and free tiers hide their commercial-use rules.

    What you get

    • Azerbaijani normalisation: numbers, ordinals (with vowel harmony), dates, time, money (manat/qəpik, $, €), percent, decimals, phone numbers, units and abbreviations become words. Brands are respelled (iPhone → Ayfon), emoji removed.
    • Letter restoration: text without ə ş ç ğ ı ö ü, typed on a Turkish keyboard or written in Cyrillic becomes correct Latin Azerbaijani, with a list of every change; doubtful words (especially names) are marked ?.
    • 10 original named style presets (5 female, 5 male): Aysel (friendly guide), Lalə (energetic Reel), Elvin (calm teacher), Nərmin (soft lifestyle), Günay (elegant luxury ad), Ayla (storyteller), Tural (news anchor), Orxan (premium brand), Rauf (motivation), Ceyhun (playful vlog). Fictional style names, not real people's voices.
    • Ready-to-paste blocks for free engines, each with 3 phone-friendly steps: Gemini TTS (Google AI Studio), ElevenLabs v3, Azure SSML (az-AZ Banu / Babek).
    • Reel timing: word budget for 15/30/60 s, a cut version if too long, scene timings per line.
    • Long texts split into chunks under each engine's limit, with a duration per chunk.
    • Orthoepy-aware stress and question intonation, plus a listening checklist with a fix for each problem (ə/e, ı/i, q/k, brands, pauses, pace).
    • Honest rules: says which free tiers forbid commercial use, adds an AI-voice disclosure line, refuses to clone real people's voices without consent.
    • Optional audio file: if your agent can run Node with network access and you set your own free GEMINI_API_KEY, the bundled scripts/tts.mjs makes the WAV/MP3. It asks before sending text to Google. Otherwise you paste the block into AI Studio yourself.

    Answers in your language (Azerbaijani by default), phone-short. Works on Claude Free and paid plans.

    Start with: "Bu mətni qadın səsi ilə 30 saniyəlik Reel üçün hazırla: …"

    Engine limits and terms change (checked 2026-11-09), so re-check before commercial use. Gemini's Azerbaijani support is in preview and its free daily limit is unpublished. Which ElevenLabs stock voices sound best in Azerbaijani varies, so test 2–3. Normalisation follows standard literary Azerbaijani. Not a substitute for a professional voice actor on high-stakes work. It never clones a real person's voice.

    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 any payment.

    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

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 2 days ago

    • Free with an Agensi account

    Requires

    Google Account

    Permissions

    Terminal / Shell
    Write Files
    Network Access
    Environment Variables

    Allowed Hosts

    aistudio.google.com

    Network access to generativelanguage.googleapis.com and the GEMINI_API_KEY environment variable are used only when the user asks for an audio file. scripts/tts.mjs reads only the user's own Gemini key, sends it only in the request header to Google, and never prints or stores it. The skill asks before sending text to Google. Everything else is text output with no network or files.

    Listed2 days ago

    What's inside

    Frequently Asked Questions

    It always gives the TTS-ready text and a ready-to-paste block for Gemini (Google AI Studio), ElevenLabs v3 or Azure. If your agent can run Node with network access and you set your own free Gemini key as GEMINI_API_KEY, the bundled scripts/tts.mjs can also make the WAV/MP3. It asks before sending text to Google and never claims audio it did not make.

    Yes. It restores ə ş ç ğ ı ö ü, converts Cyrillic to Latin Azerbaijani and lists every change. Doubtful words, especially names, are marked with ? so you can check them.

    The skill tells you per engine. For example, ElevenLabs Free does not allow commercial use, while Azure F0 does. It never says a free tier allows commercial use unless that engine's terms do, and it adds an AI-voice disclosure line for social posts.

    No. It refuses to clone or imitate a real person's voice without consent. It offers the closest of its 10 fictional style presets instead, or legal cloning of your own voice.

    Yes, on Free, Pro, Max and Team. No web search or scheduled tasks are needed. On tight Free limits it gives one engine block and one chunk per message.

    Any length. For Reels it gives a word budget for 15, 30 or 60 seconds and a cut version if the script is too long. Long texts are split into chunks under each engine's limit, with a duration for every chunk.

    New in Writing & Content