Aiko

Speech-To-Text

Aiko

Transform Your Audio into Text with Aiko

Average rated: 0.00/5 with 0 ratings

Favorited 0 times

Rate this tool

About Aiko

Introducing Aiko, an AI-powered audio transcription app that redefines the way you convert speech to text, all on your very own device. Aiko is a revolutionary application that blends the cutting-edge technology of OpenAI's Whisper model with the privacy and convenience of local processing, ensuring that your recordings never leave your device. With Aiko, experience unmatched audio transcription quality for your meetings, lectures, or any spoken content, now effortlessly transcribed into text form. Plus, with support for over 100 languages, Aiko breaks down communication barriers, making it an invaluable tool for everyone, everywhere. Aiko is not just about transcribing audio; it's about offering a seamless, user-friendly experience that caters to your privacy concerns and technical requirements. By leveraging the Whisper large v2 model on macOS and the medium or small model on iOS depending on available memory, Aiko ensures optimal performance without compromising on transcription accuracy or your device's efficiency. Moreover, its ability to transcribe audio and video files, support for various export formats like JSON, CSV, and subtitles, and integration with Siri Shortcuts for automated recording and transcription make Aiko an indispensable asset for professionals, students, and anyone in need of accurate transcription services. But what truly sets Aiko apart is its commitment to user privacy and convenience. Your audio recordings are transcribed directly on your device, keeping sensitive content secure and private. Coupled with tips for improving transcription quality, like fixing missing punctuation and dividing text into paragraphs via ChatGPT, and the ability to start recording with just a tap on your home screen, Aiko is not only smart and powerful but also incredibly intuitive and easy to use. Embrace the future of transcription with Aiko, where high-quality, privacy-centric audio transcriptions are a tap away.

Key Features

  • On-device audio transcription ensuring privacy
  • Supports transcription in over 100 languages
  • Utilizes OpenAI's Whisper model for high-quality transcription
  • Seamless integration into productivity workflows with support for shortcuts
  • Exports transcriptions to various formats (JSON, CSV, subtitles)
  • Adapts the model's size based on device memory for optimal performance
  • High privacy with direct device processing
  • Supports audio and video file transcription
  • Designed for iOS and macOS devices
  • Does not support text editing within the app

Tags

AIaudio transcriptionspeech to textprivacyOpenAI WhispermultilingualmeetingslecturesproductivitymacOSiOS

FAQs

How do I submit a feature request or report a bug?
You can submit feature requests, bug reports, or other feedback through the contact form on the Aiko webpage.
Why isn't the large v3 model used for the Mac app?
The v3 model was found to have inferior quality in many instances compared to v2. After feedback, it was decided to revert to using the large v2 model for better performance.
Can the large model be included on iOS?
The latest iPhone models lack the necessary power to run the large model efficiently. This may change with future support for multiple languages by the Whisper Distilled project.
Is it possible to edit text within the app?
Editing is not supported within Aiko. Users should export their transcription to edit it in a dedicated text editor.
How does Aiko compare to Apple's built-in transcription?
Aiko offers significantly better accuracy, supports more languages, and allows for the transcription of both audio and video files. It also supports exporting to various formats like JSON, CSV, and subtitles.
What should I do if I find mistakes in the transcription?
Since the app relies on the OpenAI Whisper model, any quality issues are outside the developer's control. However, users can provide feedback about transcription errors.
Can Aiko support more languages?
The set of supported languages is determined by the Whisper model and is not under the control of Aiko's developers. You may request additional languages from the model developers.
Why does the transcription repeat itself?
Repetitions are a known flaw of the Whisper model and are beyond the control of Aiko's development.
Why is punctuation missing from transcriptions?
Missing punctuation is a recognized limitation of the Whisper model. Aiko suggests workarounds using settings and external tools for correction.
Why does the transcription include sentences not in the audio?
Inclusion of non-audible sentences is a flaw within the Whisper model and is not something that can be adjusted by Aiko.