Today we're excited to announce a highly requested feature: manual language selection. While automatic language detection works great for most cases, we know there are times when you need more control over the transcription process.
Previously, InstantTranscriber automatically detected the language of your audio file. Now, you can manually choose from our extensive list of 90+ supported languages before starting your transcription.
When Manual Language Selection Helps
This new feature is particularly valuable in several scenarios:
- Low-quality audio - When background noise or poor recording conditions make automatic detection challenging
- Quiet or music-only intro - If the beginning has silence or only music, auto-detection may be uncertain; selecting the language (or trimming the intro) improves accuracy
- Language misclassification - If our system incorrectly identifies the language, you can now override it
- Rare or regional languages - For less common languages or specific dialects that benefit from explicit selection
- Mixed-language content - When you know the primary language but want to ensure it's prioritized
- Accented speech - Heavy accents can sometimes confuse auto-detection
90+ Languages Supported
InstantTranscriber supports a comprehensive set of languages. You can let the app auto-detect the language or choose one explicitly. Full list:
- English
- German
- French
- Spanish
- Italian
- Portuguese
- Dutch
- Swedish
- Norwegian
- Norwegian Nynorsk
- Danish
- Finnish
- Polish
- Czech
- Slovak
- Slovenian
- Hungarian
- Romanian
- Bulgarian
- Croatian
- Serbian
- Bosnian
- Russian
- Ukrainian
- Greek
- Turkish
- Arabic
- Hebrew
- Persian
- Urdu
- Hindi
- Bengali
- Tamil
- Telugu
- Kannada
- Malayalam
- Marathi
- Gujarati
- Thai
- Vietnamese
- Indonesian
- Malay
- Chinese
- Japanese
- Korean
- Cantonese
- Afrikaans
- Amharic
- Assamese
- Azerbaijani
- Bashkir
- Belarusian
- Tibetan
- Breton
- Catalan
- Welsh
- Estonian
- Basque
- Faroese
- Galician
- Hausa
- Hawaiian
- Haitian Creole
- Armenian
- Icelandic
- Javanese
- Georgian
- Kazakh
- Khmer
- Latin
- Luxembourgish
- Lingala
- Lao
- Lithuanian
- Latvian
- Malagasy
- Maori
- Macedonian
- Mongolian
- Maltese
- Burmese (Myanmar)
- Nepali
- Occitan
- Punjabi
- Pashto
- Sanskrit
- Sindhi
- Sinhala
- Shona
- Somali
- Albanian
- Sundanese
- Swahili
- Tajik
- Turkmen
- Tagalog
- Tatar
- Uzbek
- Yiddish
- Yoruba
Note: use base language names/codes (e.g., Portuguese, not pt-BR). Cantonese is listed separately from Chinese.
How It Works
The process is simple and intuitive:
- Upload your audio file as usual
- Choose your language from the dropdown menu (or leave it on "Auto-detect")
- Start transcription - our system will use your selected language for optimal accuracy
The language selector is conveniently located right next to the upload area, so you can make your choice before processing begins.
Still Automatic by Default
Don't worry - if you're happy with automatic detection (and most users are), nothing changes. Auto-detection remains the default option and continues to work excellently for clear audio in common languages.
Manual selection is there when you need it, giving you the control to ensure the best possible transcription accuracy for your specific content.
Perfect for Research and Professional Use
This feature is especially valuable for:
- Academic researchers working with international interview subjects
- Content creators producing multilingual content
- Business professionals transcribing calls with international colleagues
- Media companies handling diverse audio content
- Legal professionals working with witness testimony in various languages
As always, speaker identification and timestamps work seamlessly with manual language selection, giving you complete, accurately formatted transcripts regardless of which language you choose.
Try out the new manual language selection feature on your next transcription and let us know how it improves your workflow!