Overview
Adara Translate uses OpenAI Whisper to convert spoken audio into text. Whisper supports dozens of languages and delivers high accuracy even with background noise, accents, and varied recording conditions.
Supported Audio Formats
- MP3 — The most common audio format, widely supported.
- WAV — Uncompressed audio for maximum quality.
- M4A — Apple's audio format, common from iOS recordings.
- WEBM — Web-native format from browser recordings.
- OGG — Open-source format used in various applications.
How to Transcribe
Navigate to the Audio Transcription page. Upload your audio file or record directly in the browser. Select the language of the audio (or use auto-detect). Click Transcribe and wait for processing.
Language Support
Whisper can transcribe over 50 languages with varying accuracy. English, Spanish, French, German, Chinese, Japanese, and Korean have the highest accuracy. For best results with other languages, specify the language manually rather than relying on auto-detection.
Accuracy Tips
- Clear audio — Minimize background noise and use a good microphone.
- Single speaker — Transcription accuracy is highest with one speaker at a time.
- Moderate pace — Very fast speech or heavy overlap reduces accuracy.
- Specify language — Manual language selection avoids misdetection errors.
After Transcription
The transcribed text appears in the interface and can be copied, downloaded as a text file, or sent directly to the translator for immediate translation into another language. This workflow is ideal for creating multilingual subtitles or meeting notes.
Credit Cost
Transcription is charged at approximately 2 credits per minute of audio. A 30-minute recording costs roughly 60 credits. Check your balance before uploading long recordings.