Is my audio uploaded?
No. Audio decoding and transcription run in your browser, and your recording is not uploaded to a server. An internet connection is needed to download models and processing resources when they are not cached.
Convert speech to text for free, as often as you like. Choose an audio file or record your microphone, then create a transcript and subtitles locally.
Drop an audio file here, or click to select
Your audio is processed in this browser and is not uploaded to a server.
Standard is faster and uses less memory. Enhanced may improve recognition but needs a larger download, more memory and more time.
Auto-detect is the default. Select a language if detection is inaccurate.
Speaker numbers are estimates; review them.
Review the transcript, timestamps, and speaker labels before use.
Use this audio to text tool for speech transcription of voice notes, interviews, lectures, and meeting recordings. The first run downloads the required models; later runs can reuse the browser cache when available. Processing speed depends on your device. Review the words, timing, and any speaker labels before sharing the result.
No. Audio decoding and transcription run in your browser, and your recording is not uploaded to a server. An internet connection is needed to download models and processing resources when they are not cached.
Auto-detect identifies the language from speech near the beginning of the recording. You can also select it manually. Many languages are available, but accuracy varies with language, accent, noise, and recording quality. Mixed languages and overlapping speech may need extra review.
Enhanced is selected by default. It may improve recognition, but needs a larger first download, more memory and usually more processing time. Choose Enhanced when recognition quality is your priority. Standard is faster and uses less memory, so choose it for long recordings or devices with limited resources.
Speaker labels are optional and off by default. They estimate who spoke and can be wrong, especially for brief or overlapping voices.
Related tools: AI Audio Noise Reducer · Audio Splitter · Subtitle Editor · SRT Time Shifter · VTT Time Shifter