Transcription Details

Please note that the transcription feature is available only when the app was downloaded from the Goole Play Store.


Each transcription is on-device, which means that all processing is done on the user's device. No data is sent to servers, ensuring the privacy and security of your transcription. You can use transcription with confidence, knowing that your information is kept private and secure.


Please note that on some devices, the transcription process may not run in the background as the system blocks excessive background operations - To ensure that the transcription process runs smoothly, we recommend keeping the app open. Additionally, the transcription speed depends on the performance of your device's processor. On some devices, the only way to ensure optimal performance may be to use the "performance" variant.


  1. Record in a quiet environment: Find a quiet room or space to record your audio. Avoid noisy environments or areas with lots of background noise, such as cafes, busy streets, or near loud appliances.
  2. Speak clearly and enunciate: Speak clearly and at a moderate pace, enunciating your words properly. Avoid speaking too quickly or slurring your words together, as this can make it difficult for the transcription software to pick up what you're saying.
  3. Keep recordings short: To avoid long transcription times, try to keep your recordings short and focused. If you have a lot of material to cover, consider breaking it up into smaller, more manageable sections that can be transcribed separately.
  4. Optionally use a lossless audio format with a 16 kHz sample rate:Lossless audio formats like WAV or FLAC provide high-quality recordings that are a little bit better suited for transcription than compressed formats like MP3. Additionally, using a sample rate of 16 kHz provides the best results with a balance between recording quality and file size.

English Specific Recommendation

If you're transcribing to English, we recommend using Enhanced English dataset for optimal accuracy. This dataset has been specifically designed to recognize and transcribe English speech, and using it will help to ensure the best possible transcription results. If you're transcribing in other languages, multi-language dataset should work well, but using a language-specific dataset may improve accuracy even further.

Transcription supported languages and dialects

Albanian, Arabic, Armenian, Azerbaijani, Basque, Bengali, Bulgarian, Cantonese, Catalan, Chinese, Croatian, Czech, Danish, Dutch, English, Estonian, Finnish, French, Galician, Georgian, German, Greek, Hebrew, Hindi, Hungarian, Icelandic, Indonesian, Irish, Italian, Japanese, Kannada, Kazakh, Korean, Kurdish, Latin, Latvian, Lithuanian, Luxembourgish, Macedonian, Malay, Malayalam, Maltese, Maori, Mongolian, Nepali, Norwegian, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Serbian, Sinhala, Slovak, Slovenian, Spanish, Swahili, Swedish, Tamil, Telugu, Thai, Tibetan, Turkish, Ukrainian, Urdu, Vietnamese.