A simple, browser-based applications for transcribing audio files using the Speechmatics API or AssemblyAI API. Both companies have leading models for speech2text including speaker labelling (diarization).
- Record your meeting or conversation with client on phone
- Upload recording and transcribe audio files using Speechmatics API or AssemblyAI API
- Support for multiple languages (Czech, English, Spanish, you could simply add many others)
- Speaker diarization (identifies different speakers in transcription and label them for further processing)
- Enhanced transcription quality
- Download transcripts as text files
- Encrypted API key storage (optional local storage)
- Clean, responsive user interface
- Download the HTML file
- Open the HTML file in your web browser
- Enter your API key. API Key is stored locally but encrypted.
- Upload an audio file
- Select the language of the audio
- Click "Transcribe Audio"
- Download the resulting transcript
- A Speechmatics API key (sign up at Speechmatics)
- A AssemblyAI API key (sign up at AssemblyAI)
- A modern web browser (Chrome, Firefox, Safari, Edge)
- Audio files in a supported format (MP3, M4A, WAV, FLAC, etc.)
This is a standalone HTML file with embedded JavaScript and CSS. No server setup is required.
- Get your API key from Speechmatics
- Get your API key from AssemblyAI
- Open the HTML file in any web browser
- Enter your API key in the designated field
- Your API key is stored encrypted in your browser's local storage if you select "Remember API key"
- All processing is done via the Speechmatics API or AssemblyAI API; no audio data is stored elsewhere
- The application works entirely in your browser
- Transcription quality depends on the Speechmatics or AssemblyAI models
- Large audio files may take longer to process
- The free tier of Speechmatics API has usage limits (4 hours each month)
- The free tier of AssemblyAI give you one time credit $50
MIT License
Feel free to contribute to this project by submitting pull requests or reporting issues.