SingLang Manual

🚀

What is SingLang?

SingLang is a browser-based language learning tool that converts any phrase into a sung translation. You type something in your own language, choose a target language and a musical genre, and an AI singer performs your phrase — translated and sung — within seconds.

The idea is simple: music aids memorization. Rhythm, melody, and repetition are among the most powerful tools for retaining new vocabulary. SingLang combines AI translation (Claude), AI vocal generation (ElevenLabs), and structured learning tools (vocabulary cards, flashcards, karaoke, quiz) into a single tool that runs entirely in your browser.

Who is SingLang for?

• Language learners who want an alternative to dry vocabulary drills
• Music fans curious to hear their favorite phrases in other languages
• Teachers looking for engaging classroom tools
• Travelers who want to memorize key phrases before a trip
• Anyone who thinks learning with ears is more fun than learning with eyes

⚡

Quick start

Getting your first sung translation takes about 30 seconds:

1. Open singlang.com in your browser (desktop or mobile)
2. Type a phrase in the text box — up to 100 characters, in any language
3. Click Translate & Speak for a free instant spoken translation
4. For a sung version: choose a target language and a rhythm, then click Translate & Sing (requires credits)
5. Listen to the result, read the pronunciation guide, and explore the vocabulary cards

Try it free — no credit card

The Translate & Speak button uses Google's translation + TTS API and is completely free, unlimited, and instant. It gives you the translated text, pronunciation, and vocabulary. Use it as many times as you like without any credits.

🖥️

The interface

SingLang's app is a single page with all controls visible. From top to bottom:

• Header: Logo, share button, and light/dark mode toggle
• Text input: Where you type your phrase (100 character limit)
• Language grid: 20 flags to choose the target language
• Rhythm grid: Musical genre selector — 12 always visible, 84+ behind "More rhythms"
• Voice type: Male, Female, or Child (Basic+ plans)
• Quiz mode toggle: Activates fill-in-the-blanks mode (Basic+ plans)
• Action buttons: Translate & Speak (free) | Translate & Sing (credits)
• Result area: Translation, pronunciation, karaoke, vocabulary, flashcards
• History & Favorites: Recent translations and saved songs

🔊

Translate & Speak Free forever

The Translate & Speak button provides instant translation and spoken audio using Google's services. It is completely free, requires no credits, and has no usage limit.

What it provides

• The translated text in your chosen language
• Spoken pronunciation via Google TTS
• Detected source language

Example audio demos

Here are simulated examples of what Translate & Speak produces (click ▶ to hear in the actual app):

🎵

Translate & Sing Requires credits

The core feature of SingLang. Click Translate & Sing to generate a full AI vocal performance of your phrase in the target language and chosen rhythm. Each generation consumes 1 credit.

The 3-step process

1. Translation (Claude AI): Your phrase is sent to Claude with musical and cultural context. The AI adapts the translation to be singable — preserving meaning while fitting the rhythm and tone you selected.
2. Vocal generation (ElevenLabs): The translated text is sent to ElevenLabs' Music API with a detailed prompt describing the genre, voice type, language, and style. An AI vocalist performs it.
3. Lyric sync (ElevenLabs STT): The generated audio is analyzed by ElevenLabs' speech-to-text API to extract word-level timestamps, enabling karaoke-style highlighting.

Example sung demos

🗣️

Pronunciation Guide Always included

Every translation — whether spoken or sung — includes a pronunciation guide in the result panel. This is generated by Claude and provides a phonetic rendering in the Latin alphabet, written to be readable by English speakers.

Example

Phrase: "The sky is beautiful tonight" → Japanese: 今夜の空は美しい

Pronunciation: Kon-ya no so-ra wa u-tsu-ku-shi-i

This appears below the main translation, marked with a purple left border. For languages with non-Latin scripts, the romanization block (see below) adds an additional layer of detail.

📚

Vocabulary Cards Always included

Below the translation result, a vocabulary table appears automatically. It extracts 3–6 key words from the translated phrase and displays:

• Target word — the word in the target language
• Pronunciation — how to say it phonetically
• Meaning — the English equivalent

These cards are the source for the Flashcard and Quiz features. The words chosen are the most useful or interesting from the phrase — not every word, but the ones worth remembering.

🔤

Romanization Credit 35+

For languages that use non-Latin scripts, SingLang automatically adds a romanization block — a full phonetic transcription of the translated text in the Latin alphabet. This is separate from the pronunciation guide and covers the entire translated sentence.

Supported languages

🃏

Flashcards Starter+

After any sung translation, click the 🃏 Cards button in the action row to open the flashcard session for that song's vocabulary.

How flashcards work

1. The front of the card shows the word in the target language
2. Tap the card to flip it and reveal the pronunciation and meaning
3. Rate yourself: ❌ Didn't know or ✅ Got it!
4. Work through all cards — your score is shown at the end

Flashcard TTS (Basic+)

On Basic plan and above, when you flip a flashcard, the word is automatically spoken aloud using the Translate & Speak API. This reinforces the correct pronunciation at the moment of recall — a powerful combination of visual and auditory learning.

🎤

Karaoke Mode Starter+

Karaoke mode displays the translated lyrics and highlights each word in real time as the AI singer performs it. It requires a successful lyric sync (step 3 of generation) to activate.

How to use

1. After generation completes, click 🎤 Karaoke in the action row
2. Press Play on the audio player
3. Watch each word light up in sync with the singing
4. Words turn blue once sung, active word turns white with a purple background

Technical notes

Karaoke sync uses ElevenLabs' Automatic Speech Recognition API to extract word-level timestamps from the generated audio. This process adds about 5–10 seconds to generation time. If the AI singer adds musical flourishes or repeats syllables, the sync may be slightly approximate — this is normal and doesn't affect the learning value.

🎯

Quiz Mode Basic+

Quiz Mode transforms every sung translation into an active recall exercise. Instead of reading the full translation, you listen to the song and fill in the blanks — then check your answers.

How to activate

1. Toggle Quiz Mode to "On" before clicking Translate & Sing
2. The translation text will be hidden; a fill-in-the-blanks sentence appears instead
3. Listen to the song (you can replay it multiple times)
4. Type the missing words into the input fields
5. Click ✓ Check to see your score, or 👁 Reveal to see answers

Which words become blanks?

Claude selects 2–4 vocabulary words to hide as blanks — typically the most meaningful or interesting words, not function words like "the" or "and". This ensures the quiz tests real vocabulary, not grammar particles.

🎶

Pulsing Lyrics Advanced+

The 🎵 Lyrics button opens a pulsing lyrics display: the full translated text in large format, with a subtle scale animation on each beat detected in the audio.

Beat detection uses the Web Audio API's AnalyserNode to calculate RMS energy from the audio stream in real time. When energy crosses a threshold, the text pulses. It's a visually engaging way to stay focused on the lyrics while the music plays.

🌍

Languages

SingLang supports 20 target languages for translation and sung generation:

SingLang detects the language of your input automatically — you can type in any language and it will translate to whichever target language you've selected. Right-to-left languages (Arabic, Hebrew) are displayed and sung with correct directionality.

🥁

Rhythms & Genres

SingLang includes 100 musical genres and rhythms, ordered by global popularity. The first 12 are always visible; the rest expand with "▼ More rhythms".

Always visible (all plans)

Genre access by plan

Genres beyond your plan limit appear dimmed. Clicking them shows an upgrade prompt. The full genre list includes global styles from Bolero, Guarânia, and Fado to Bhangra, Gnawa, and Gregorian chant.

🎙️

Voice Types Basic+

Choose from three vocal character types before generating your song:

• 👨 Male — deep masculine tenor voice
• 👩 Female — warm clear soprano voice
• 👦 Child — young sweet voice, kids vocalist style

Voice type is passed to ElevenLabs as part of the generation prompt, combined with the language and rhythm descriptors. On free plans and Credit 12, voice cards appear dimmed and clicking them opens the upgrade modal.

🎨

Custom Rhythm Premium+

Premium and VIP subscribers can describe their own musical style in free text. A text input appears below the rhythm grid — type anything like:

• "Slow jazz with ukulele and light rain sounds"
• "80s power ballad with electric guitar solo and reverb chorus"
• "Celtic lullaby with tin whistle and soft bodhrán"
• "Minimalist piano, just one note at a time, very slow"

Your description is combined with the voice type and language into the ElevenLabs generation prompt. The more specific you are, the closer the result to your vision — though AI generation always adds its own interpretation.

💳

Credits & Plans

Each Translate & Sing generation costs 1 credit. Credits are deducted only on successful generation — if the process fails, your credit is returned.

Plan comparison

Session IDs and credits

SingLang doesn't require account creation. Your credits are tied to a session ID stored in your browser's localStorage. This ID is also stored server-side, so your credits persist across page reloads and browser restarts — as long as you don't clear localStorage.

Payments

All payments are processed securely by Paddle, a Merchant of Record. This means Paddle handles taxes, VAT compliance, and payment security. You'll see Paddle on your bank statement. All major credit cards and PayPal are supported.

🕐

History & Favorites Basic+ for favorites

History

Your last several translations are automatically saved in the History section at the bottom of the page. Click any history item to reload its phrase into the input field. History limits by plan:

• Free / Credit 12 / Credit 35 — last 5 translations
• Starter — last 10 translations
• Basic+ — last 100 translations (effectively unlimited)

Favorites

On Basic+ plans, the ⭐ Save button saves the current translation AND its audio file to your browser's IndexedDB. Favorites persist permanently (until cleared) and include a play button to replay the audio without generating again.

This is useful for building a personal library of phrases you want to practice — you can replay your saved songs without spending any credits.

❓

Frequently Asked Questions

✨

Pro Tips

• Use Ctrl+Enter (Cmd+Enter on Mac) to trigger generation without clicking the button.
• Short phrases work best. 5–15 words generate more natural-sounding music than very long sentences.
• Emotionally resonant phrases tend to produce better musical output — phrases that have inherent rhythm and feeling.
• Repeat listening. Research on music and memory shows that 3–5 listens to a melodic phrase significantly improves retention compared to reading the same phrase once.
• Use Quiz Mode after 2 listens, not before. Let the melody carry the words in first.
• Try the same phrase in different rhythms. Hearing "I miss you" as a Bossa Nova vs. a K-Pop track makes different words stick differently.
• Combine with flashcards: Generate → listen 2× → flashcards → listen 1× → quiz. This cycle takes about 3 minutes per phrase and covers listening, reading, recall, and testing.
• Save your best generations to Favorites (Basic+). Build a personal "playlist" of phrases you want to keep practicing — replaying saved audio costs no credits.
• On mobile: Use headphones for the karaoke experience. The word highlighting is much more engaging when you're fully immersed in the audio.