Speech Text is an online AI speech to text workspace that turns audio, video, live browser recordings, and media URLs into editable, searchable transcripts. Powered by OpenAI Whisper, it supports more than 100 languages and runs entirely in the browser with no setup.
What is Speech Text?
Speech Text is a browser-based speech to text tool that transcribes audio and video into text you can search, edit, and export. It accepts file uploads (MP3, WAV, M4A, MP4, MOV, WEBM), live browser recordings, and media URLs, and produces transcripts with timestamps and speaker labels. The service uses OpenAI Whisper technology and processes transcription in the browser via WebGPU, Transformers.js, and ONNX Runtime.
Key Features
- Multiple input methods — Upload audio or video files, record live in the browser, or paste a supported media URL.
- OpenAI Whisper transcription — Uses OpenAI Whisper, a multilingual model that handles varied accents and real-world recording conditions.
- 100+ language support — Automatic language detection with manual override across over 100 languages, from English and Spanish to Cantonese and Luxembourgish.
- Speaker labels and timestamps — Transcripts include speaker labels and timestamps, useful for interviews, meetings, and panels.
- Transcript editor with search — Search and edit the transcript directly in the browser, then prepare it for publishing or internal notes.
- Export formats — Download as TXT, SRT, VTT, DOCX, JSON, or PDF for captions, documents, searchable archives, and structured data.
- Pro AI tools — Paid plans add AI Summary, AI Analytics, Chat with AI, and translation into 100+ languages.
Who should use Speech Text?
Journalists can transcribe interviews and convert them into searchable notes. Podcasters can turn episode audio into drafts for show notes, captions, or blog posts. Video editors can create SRT subtitle files from video tracks. Students and researchers can convert lectures into searchable transcripts for studying.
What can you do with Speech Text?
- Meeting notes — Record a live meeting in the browser and get an editable transcript with speaker labels for action items.
- Interview transcription — Upload an interview recording, select the language, and export the transcript as DOCX or JSON for documentation.
- Subtitle generation — Convert video to SRT subtitles by uploading an MP4 or MOV file and exporting in SRT or VTT format.
- Podcast repurposing — Turn a podcast episode into text for show notes, searchable archives, or AI summaries, then translate it into other languages.
How does Speech Text work?
Start a transcription job by uploading a file, recording live, or pasting a media URL. The audio is processed in the browser using OpenAI Whisper, and the transcript appears in an editor where you can search, correct, and add speaker labels. When ready, export the transcript in TXT, SRT, VTT, DOCX, JSON, or PDF format.
Pricing
Speech Text uses a freemium model. New users receive 5 free transcription minutes. Paid plans offer yearly billing with monthly equivalents:
- Starter — $4.90 per month (billed yearly at $58.80) includes 1,440 transcription minutes per year, uploads up to 1GB, speaker labels, transcript editor, and exports in TXT, SRT, VTT, JSON, PDF, and DOCX.
- Pro — $14.90 per month (billed yearly at $178.80) includes 7,200 minutes per year, plus AI Summary, AI Analytics, Chat with AI, translation, and email delivery.
- Max — $24.90 per month (billed yearly at $298.80) includes 36,000 minutes per year, the lowest per-minute cost, private storage for audio, and all Pro features.
FAQ
Is Speech Text free?
Yes, Speech Text gives new users 5 free transcription minutes to test the workflow. After that, you can subscribe to Starter, Pro, or Max plans, or buy credit packs for extra transcription minutes.
What languages does Speech Text support?
Speech Text supports over 100 languages, including English, Spanish, Chinese, French, German, Portuguese, Russian, Korean, Japanese, and many more. It can auto-detect the language or use your manual selection.
Can I use Speech Text to transcribe videos?
Yes. You can upload video files in MP4, MOV, or WEBM format, or paste a media URL. Speech Text extracts the spoken track and converts it into an editable transcript, which you can export as SRT subtitles or other formats.
What export formats are available?
Speech Text exports transcripts as TXT, SRT, VTT, DOCX, JSON, and PDF. This covers plain text, subtitles, documents, and structured data for downstream applications.
How accurate is Speech Text transcription?
Accuracy depends on source quality. Clear audio with minimal background noise, single speakers, and standard vocabulary produces the best results. The page notes that important names and facts should always be reviewed.
