Accuracy
Consistently delivering low word error rate across all languages and use cases. Frequent testing of our language offering positions us ahead of the biggest players in the market and ensures continuous innovation to deliver the best accuracy for every use case.
Language Coverage
An industry-leading language coverage ensures our offering is relevant to your business needs. Find out more about our language offering.
Flexible Deployment
Consume Speechmatics’ speech-to-text technology in real-time or with pre-recorded (batch) files, in the cloud or securely on-premises. Speechmatics has the flexibility to be deployed according to your needs. Learn more about our robust and scalable deployment options.
Advanced Punctuation
Speechmatics’ Advanced Punctuation is built on over 2.5 billion words and has an industry-leading set of supported punctuation marks. This use of punctuation optimises the speed and ease of consuming a transcript for human users.
Custom Dictionary and Sounds Feature
Add a set of context-specific words to the dictionary to enhance your transcription accuracy. Learn more about the benefits of Custom Dictionary.
Speaker Change
Easily identify a change of speaker within your transcript with Speaker Change. A token is automatically added to the transcript each time a speaker change is detected. This enables easy modification of the transcript to improve its readability.
Confidence Scores
Visualise the confidence of every word within the transcript.
Speaker Diarisation
Detect and label different speakers within the same channel.
Channel Diarisation
Detect and label speakers on up to 6 streams or channels.
Extensive file format support
Accelerate your transcript turnaround time by reducing the time it takes for you to prepare your audio or video files. Speechmatics supports an extensive set of file formats so you don’t have to worry about converting files to suit our requirements.
Automatic Sample Rate Detection
Speechmatics’ ASR can determine the sample rate of each media file and applies the most appropriate transcription model to optimise accuracy.
Low Latency Finals
Low Latency Finals enable the most accurate real-time transcription by leveraging Speechmatics’ proprietary rescoring outputs. The feature is able to define the context of a transcript and automatically correct words to match the context.
