Speechmatics offers a range of different products and speech recognition services. Using a completely new deep learning approach we are achieving world beating accuracy levels.
Our time aligment service automatically synchronises text with audio or video files. Simply upload your transcript and media files to receive a timestamped document in any language.
We are at the cutting-edge of deep and recurrent neural network algorithms. Our ground-breaking approach, pioneered by Dr Tony Robinson, is improving accuracy continuously and pushing the boundaries of what was thought possible in speech recognition.
We operate a very easy to use cloud-based model through a simple REST API that reduces your up-front costs and helps maximise returns. We also deploy on premises solutions to selected clients.
Register with Speechmatics today and get a completely free trial of audio transcription and / or alignment! No strings attached.
Support for .mp3 .mp4 .avi .wav .flac .ogg and more. Transcribe an ever increasing range of languages using our different language products. Automatic punctuation, capitalisation, diarisation (speaker separation) and individual word timings and confidences.
Auto-Auto is a ground-breaking new framework used by Speechmatics to build languages for use in Automatic Speech Recognition. This technology allows us to rapidly create and continuously improve any language or variants of existing languages. We are constantly adding support for new languages or specific use-cases such as regional dialects, industry domains or accents.
Vote For New Languages
Arabic
German
Australian English
British English
North American English
Spanish
French
Italian
Dutch
Polish
Portuguese
Swedish
Russian
Submit new jobs and monitor progress of running jobs.
Download previous transcriptions to .txt, .json or a print friendly version.
Select which language product you want to use.
Top up credit easily with Stripe.
View transcript with confidences and timings of each word.