Aiko 4+
AI Speech-to-Text (Transcribe)
Sindre Sorhus
-
- Free
Screenshots
Description
High-quality on-device transcription. Easily convert speech to text from meetings, lectures, and more.
The transcription is powered by OpenAI's Whisper running locally on your device. Nothing leaves your device.
You can export the transcription as subtitles too.
Aiko favors accuracy over speed.
The app requires a Mac with at least 16 GB of RAM.
Supports audio in 100 different languages:
Afrikaans, Albanian, Amharic, Arabic, Armenian, Assamese, Azerbaijani, Bangla, Bashkir, Basque, Belarusian, Bosnian, Breton, Bulgarian, Burmese, Catalan, Chinese, Croatian, Czech, Danish, Dutch, English, Estonian, Faroese, Finnish, French, Galician, Georgian, German, Greek, Gujarati, Haitian Creole, Hausa, Hawaiian, Hebrew, Hindi, Hungarian, Icelandic, Indonesian, Italian, Japanese, Javanese, Kannada, Kazakh, Khmer, Korean, Lao, Latin, Latvian, Lingala, Lithuanian, Luxembourgish, Macedonian, Malagasy, Malay, Malayalam, Maltese, Marathi, Mongolian, Māori, Nepali, Norwegian, Norwegian Nynorsk, Occitan, Pashto, Persian, Polish, Portuguese, Punjabi, Romanian, Russian, Sanskrit, Serbian, Shona, Sindhi, Sinhala, Slovak, Slovenian, Somali, Spanish, Swahili, Swedish, Tagalog, Tajik, Tamil, Tatar, Telugu, Thai, Tibetan, Turkish, Turkmen, Ukrainian, Urdu, Uzbek, Vietnamese, Welsh, Yiddish, Yoruba
The app was made possible thanks to Whisper by OpenAI and whisper.cpp by Georgi Gerganov.
■ FAQ
‣ Can I edit the text in the app?
I don't plan to support any editing. Export the transcription and edit it in a proper text editor.
‣ Why is the app so large?
The app delivers the highest quality transcription on the market for 100 different languages. Rather than asking why it's so large, the real question is how is it so small.
‣ Why is the transcription so slow?
The favors accuracy over speed. However, performance is expected to improve in the coming months.
‣ Can I delete some of the languages to save space?
This is unfortunately not possible. The model has all the languages stored together in a way that makes it impossible to remove just some languages.
More FAQs on the website.
■ Technical details
The app uses the Whisper large v2 model on macOS and the medium or small model on iOS depending on available memory.
■ Privacy
The transcription is done locally on your device. Nothing leaves your device. The app is not even able to connect to the internet due to self-imposed restrictions that are enforced by macOS (no network entitlement).
■ Support
You can contact me through the feedback button in the “Help” menu in the app.
What’s New
Version 1.7.0
- Added setting for word replacements.
Ratings and Reviews
Phenomenal and somehow free?
I have researched endlessly trying to find my options to transcribe the many discussions I recorded with my parents about their past. Now that they are no longer with us I was looking for options to transcribe so many recordings with non-ideal audio. (We’d often chat with the TV on in the background, for example) I wasn’t expecting great results but hoping for a starting point and weeding through random background noise transcriptions. Somehow these things are so good and it only grabs the conversations. I need to clean them up a bit of course because no non-human translation is perfect, but it is so much better than I was hoping! I rarely leave reviews, but I’m almost in tears with how well this has suited my needs!
The answer to my prayers
I often have to record something on my phone, then, transcribe it, I was using a different service for this, but that cost money. Every time I was using the transcriptions, and it took a while to come back. This is wonderful, it can do it on device, I knew something like this was going to come out when I saw the whisper model last year, but back, then it could only run on the computer, and this is so much better. I have both versions, the one for my Mac, and one for my phone, and they both are wonderful, one for when I’m sitting down and one for when I’m on the go.
Works with 8 GB of RAM—no need for 16 Gigs
I cannot tell how it would perform with 16 GB's, but even with 8, it was able to transcribe speech where I mixed English words with a language spoken by less than 10 million natives. It is somewhat slow with transcribing audio files (at least on 8 GB's) but the luxury to not have to use the f— cloud, and sell your private life out AND pay for it via the API fee (0.5¢/~700 words) is the cherry on the cake. Works with all sorts of file formats, you drag it, drop it, and that's about it; you get a transcript.
App Privacy
The developer, Sindre Sorhus, indicated that the app’s privacy practices may include handling of data as described below. For more information, see the developer’s privacy policy.
Data Not Collected
The developer does not collect any data from this app.
Privacy practices may vary, for example, based on the features you use or your age. Learn More
Information
- Seller
- Sindre Sorhus
- Size
- 2.9 GB
- Category
- Productivity
- Compatibility
-
- iPhone
- Requires iOS 17.4 or later and a device with the A12 Bionic chip or later.
- iPad
- Requires iPadOS 17.4 or later and a device with the A12 Bionic chip or later.
- Mac
- Requires macOS 14.4 or later.
- Apple Vision
- Requires visionOS 1.1 or later and a device with the A12 Bionic chip or later.
- Languages
-
English
- Age Rating
- 4+
- Copyright
- © Sindre Sorhus
- Price
- Free