Skip to content

AI Models

Whisper Memos lets you choose which AI model transcribes your recordings. Different models have different strengths — some are faster, some are more accurate, and some support more languages. You can switch models anytime in Settings → AI Model.

The fastest option available. OpenAI Whisper Turbo delivers near-instant results for most recordings and supports auto-detection across 99+ languages. A great choice when speed matters most.

  • Speed: ~97x real-time (a 10-minute memo transcribes in about 6 seconds)
  • Languages: 99+ with automatic detection
  • Plan: Free

The default model for all users. Whisper offers slightly better accuracy than Turbo at the cost of somewhat slower processing. If you haven’t changed your settings, this is what you’re using.

  • Speed: ~58x real-time
  • Languages: 99+ with automatic detection
  • Plan: Free

Released in March 2026, Cohere Transcribe is one of the most accurate transcription models available, topping the Open ASR Leaderboard. It supports 14 languages with particularly strong results in English, German, French, Spanish, and Japanese.

  • Speed: ~30x real-time
  • Languages: 14 — English, German, French, Italian, Spanish, Portuguese, Greek, Dutch, Polish, Arabic, Vietnamese, Chinese, Japanese, Korean
  • Plan: Free

Because Cohere doesn’t auto-detect language, Whisper Memos runs a quick language check on your recording first. If the detected language is in Cohere’s supported set, it proceeds with Cohere. Otherwise, it falls back to Whisper automatically.

ElevenLabs Scribe offers high accuracy across a very wide range of languages, including less common ones. It’s the best choice if you frequently record in languages that other models struggle with.

  • Speed: ~17x real-time
  • Languages: 99 with automatic detection
  • Plan: Pro
ModelSpeedLanguagesAccuracyPlan
OpenAI Whisper TurboFastest99+GoodFree
OpenAI WhisperFast99+GoodFree
Cohere TranscribeModerate14HighestFree
ElevenLabs ScribeSlower99HighPro

  1. Open Settings in Whisper Memos
  2. Tap AI Model & Settings
  3. Select the model you want to use
  4. All future recordings will use the new model

Switching models doesn’t affect your existing memos — only new recordings use the selected model.

  • Start with the default: OpenAI Whisper works well for most people and most languages
  • Try Cohere for supported languages: If you record in one of Cohere’s 14 languages, it’s worth trying for the accuracy improvement
  • Use ElevenLabs for rare languages: If your language isn’t well served by the other models, Scribe’s broad language support may help
  • Fallbacks are automatic: If your chosen model can’t process a recording for any reason, Whisper Memos automatically tries the next best option