Transcribe or translate
Task lets you transcribe the speech or translate it into English.
Transcribe speech
Wizper (Whisper v3)
Whisper v3 Large, optimized for speed: the same word error rate at about twice the performance.
Made by OpenAI50 creditsper run Pro
Opens the audio studio with Wizper (Whisper v3) ready to use. You can look around as a guest. Running it takes the Pro plan or higher.
What Wizper (Whisper v3) makes, what it needs and what you get
What it makes
Transcribe speech
Turns speech in a recording into written text.
What you need
- The recording to transcribe
What you get
A transcript of what was said
About Wizper (Whisper v3)
Wizper (Whisper v3) turns speech in an audio clip into text. It uses the Whisper v3 Large model. You add one audio file in mp3, mp4, mpeg, mpga, m4a, wav or webm format. You can then transcribe it or translate it into English.
The Task option has two choices, Transcribe and Translate. Language is the language of the audio and starts on English. If you choose Translate, the text comes back in English whatever language you pick. Max Segment Len sets the longest piece of speech, from 10 to 29 seconds, before the audio is split. Merge Chunks joins neighboring pieces while they stay within that length.
It costs 100 credits. Whisper costs 200 credits and takes only the audio clip.
Other models for this job
Same job: transcribe speech. Compare before you choose.
Audio Free Open
WhisperTranscribe speech
Turn audio and video into accurate, timestamped text in 99 languages. Run Fun AI studio Whisper in your browser no installs, no code
OpenAI200 creditsper run
Audio Pro Open
StreamTranscribe speech
Leverage the rapid processing capabilities of AI models to enable accurate and efficient real-time speech-to-text transcription.
Open Source40 creditsper 5 s
Audio Pro Open
Nemotron Asr MultilingualTranscribe speech
Nemotron-ASR-Streaming is a multi lingual, streaming Automatic Speech Recognition (ASR) engineered to deliver high-quality multi lingual transcription across both low-latency streaming and high-throughput batch workloads.
NVIDIA7 creditsper 5 s
Audio Pro Open
Speech To TextTranscribe speech
Leverage the rapid processing capabilities of AI models to enable accurate and efficient real-time speech-to-text transcription.
Open Source40 creditsper 5 s
What it does well
Several audio formats
Accepts mp3, mp4, mpeg, mpga, m4a, wav and webm files.
Whisper v3 Large
Uses the Whisper v3 Large model.
How to use Wizper (Whisper v3)
How It Works
How It Works
01
Add your audio file
Upload one file in mp3, mp4, mpeg, mpga, m4a, wav or webm format.
02
Choose Task and Language
Pick Transcribe or Translate, and set the language of the audio.
03
Generate the text
Run the model and read the text it returns.
Tip Language starts on English, so change it if your audio is in another language.
Good for
Transcribing a recording
Turn a recorded talk, call or interview into text.
Translating speech into English
Turn a recording in another language into English text.
Questions
Quick answers to help you get the most out of Wizper (Whisper v3) in FunAI Studio.
1. Which files can I add?
2. Can it translate?
3. What do I get back?