WhatIsTTS
Log In Sign Up
WhatIsTTS

Home / Speech to Text

Speech to Text

Transcribe audio and video files to text with AI. Supports 99 languages, multiple models, and automatic language detection.

Upload Audio or Video

Drag & drop your file here, or browse

Supports MP3, WAV, FLAC, OGG, M4A, WEBM, MP4. Max 50MB.

file.mp3

0 MB
Transcribing your audio... This may take a moment for longer files.

Transcription Result

Credit Cost

2 credits per minute of audio

Billed based on audio duration, rounded up to the nearest minute.

Available Models

OpenAI Whisper

OpenAI's robust speech recognition API supporting 57 languages.

57 languages Translation to English Timestamps Noise robust

OpenAI GPT-4o Transcribe

OpenAI's latest transcription model with improved accuracy and structured output.

Improved accuracy Structured output Better punctuation Multi-speaker handling

Deepgram Nova-3

Deepgram's fastest and most accurate STT with real-time streaming and diarization.

Real-time streaming Speaker diarization Smart formatting Topic detection Noise robust

Google Cloud Chirp 3

Google's latest speech recognition with 100+ language support and auto-punctuation.

100+ languages Auto-punctuation Speaker diarization Word confidence scores Streaming

Microsoft Azure STT

Azure speech recognition with 100+ languages, custom models, and real-time streaming.

100+ languages Custom models Real-time streaming Speaker identification Batch transcription

ElevenLabs Scribe

ElevenLabs' speech-to-text with speaker diarization and 99 language support.

99 languages Speaker diarization Timestamps Intelligent formatting Multi-speaker

Best For

Meetings, interviews, podcasts, lectures, support calls, and searchable audio archives.

Frequently Asked Questions

Speech to text (STT), also called automatic speech recognition (ASR), converts spoken language into written text. Our models use AI to accurately transcribe audio from meetings, interviews, podcasts, lectures, and more.

OpenAI Whisper is the most cost-effective choice for general use. For highest accuracy, try OpenAI GPT-4o Transcribe. Deepgram Nova-3 excels at real-time streaming. ElevenLabs Scribe is great for multi-speaker recordings with automatic diarization. Google Chirp 3 covers 100+ languages.

We support MP3, WAV, M4A, OGG, FLAC, WEBM, and most common audio/video formats. Maximum file size is 50MB. For larger files, consider splitting the audio first.

Free users can transcribe up to 5 minutes of audio. Paid plans support audio files up to 2 hours. For longer recordings, use our API with batch processing.

Our providers achieve 95%+ accuracy on clear English speech. Accuracy varies by language, audio quality, and background noise. OpenAI Whisper supports 57 languages, ElevenLabs Scribe supports 99, and Google Chirp 3 supports 100+.

Speech to text costs 2 credits per minute of audio transcribed. Free users can try the tool with limited duration. Credits are deducted based on actual audio length, not file size. Check our pricing page for credit packages and subscription plans.

Yes. ElevenLabs Scribe and Deepgram Nova-3 provide automatic speaker diarization, labeling different speakers in the transcript. This is especially useful for meetings, interviews, and multi-party recordings. Google Chirp 3 also supports diarization for supported languages.

Yes. All providers return word-level or segment-level timestamps. This enables subtitle generation, audio-text alignment, and precise navigation within long recordings. Timestamps are included in both the web interface and API responses.

Yes. Use our REST API at /api/v1/stt/ with your API key (Bearer sk-tts-...). Upload audio as multipart form data and receive JSON with the transcript, timestamps, and confidence scores. The API supports all STT providers and output formats.

Use a high-quality recording with minimal background noise. Speak clearly and at a moderate pace. Use lossless audio formats (WAV, FLAC) when possible. For noisy recordings, run our Audio Enhancer tool first to clean up the audio before transcribing.

Yes. Upload video files (MP4, MOV, WEBM, AVI) and the audio track will be extracted and transcribed automatically. The 50MB file size limit still applies. For larger video files, extract the audio track first using a tool like FFmpeg.

Audio files are uploaded securely over HTTPS and sent to the selected transcription provider for processing. Files are not stored permanently on our servers. Each provider has its own data retention policy — OpenAI and Deepgram do not use API data for training. Check each provider's privacy policy for details.

Need a specific TTS workflow?

Compare providers, test voices, then run it through one brokered API.

View Plans API Docs