Below you’ll find answers to some commonly asked questions.

Speechmatics support many audio and video file formats including wav, mp3, aac, ogg, flac, wma, mpeg, amr, caf, mp4, mov, wmv, mpeg, m4v, flv, mkv. Other formats may work but should be validated under user acceptance testing.

For best possible results try to avoid using file formats that use compression technologies.

For a full list of features including file input formats, please see our product sheets.

A full list of our available languages can be found on the languages page.

Yes. All audio/video files submitted to the Speechmatics ASR require a language code to be supplied. A list of all the language codes are provided in related customer documentation.

All Speechmatics language packs are optimised for sample rates of 8kHz and 16kHz. If you are unsure or your audio files are varied and have mixed sample rates, please submit them as is, and the Speechmatics engine will automatically optimise them for the best possible results.

Yes. Our Batch ASR solutions support audio/video files with up to 6 channels. When enabled the audio/video file submitted will have each channel processed separately but will be combined in a single output. Some benefits of Channel Diarisation:

  • Custom channel labels can be submitted for example Agent & Caller
  • Can eliminate cross talk
  • A single output is generated that has channel labelling and timing information for each word recognised
  • Perfect speaker identification based on 1 user per channel

Things we have seen impact accuracy include:

  • Clarity of speech
  • Background noise
  • Cross talk
  • Distance from microphone

Other factors that can impact include multiple languages in a single audio file.

Speech recognition is limited by the quality of the audio presented to the system. To get the best out of our system, try to follow these simple steps below.


  • Speak clearly. However, there is no need to speak slowly
  • Ideally, your voice should flow over the microphone, not directly into it for optimal results
  • Literal translation of people ‘thinking out loud’ is very hard to read, so – if possible – think about what you want to say before you say it
  • If you have time, record some test audio and play it back to check it all sounds okay before you try it live

Your environment:

  • Record in a quiet environment to minimise background noise
  • Try to avoid multiple people speaking at the same time
  • Minimise reverberation. Sound can sometimes bounce off flat walls and muddy the signal

Your technology:

  • Use a good microphone, such as a USB noise cancelling (or directional) microphone – a head mounted microphone is preferable
  • Record at 16kHz or greater if possible
  • There is no need to compress the audio, if you do, please don’t over compress – use 96 kbps AAC or 128 kbps MP3 or better.
  • Use two channels if available.
  • Do not apply any transcoding prior to submitting to the Speechmatics ASR

Our flexible approach to ASR allows you to deploy our on-premises solutions for batch and real-time use cases, providing peace of mind regarding data security and allows you to meet any data compliance requirements.

Check out the product sheets to see the various deployment options available.

We support various forms of diarisation and speaker separation to suit different use-cases. These are:

  • Speaker diarisation detects and labels the gender (male and female) speakers in an audio/video file. Available in our batch ASR solutions
  • Channel diarisation allows a single audio file with multiple tracks to be submitted to the ASR with custom labels (eg. agent and customer). Available in our batch ASR solutions

Any words in the text that are not in the audio provided need to be removed or surrounded by <>, otherwise accuracy will be reduced and the alignment job may take significantly longer.

Alignment is only available on our SaaS.

Actions speak louder than words. We like our customers to try audio files representative of their use-case. This provides a true measurement of accuracy and what it means for you. Why not try transcribing a media file or live speech using our free demo and see for yourself?

Users of our public cloud ASR can refer to our Terms & Conditions and Privacy Policy.

Speechmatics’ on-premises solutions do not collect or store any video/audio files or transcribed output, this gives you full control over your data.

Batch ASR is a transcription of pre-recorded audio and video files. A transcript is provided once an entire audio file has been processed.
Real Time ASR is a transcription provided in parallel to input audio being streamed through Speechmatics engine. Responses from the engine can be as fast as 1 second.

The use of any of our solutions will depend on your use case. For example, for real-time captioning, you may want to consider the real time engine, whereas batch processing is more suited to transcribing an archive of data.

The processing time taken to transcribe a media file compared to the length the file is referred to as the real time factor (RTF).

Our batch ASR will transcribe a file with an RTF of 0.5. For example, a 10-minute file will be transcribed within 5 minutes.

We cannot guarantee an RTF of 0.5 for files less than 3 minutes in length.

The processing time taken to transcribe a media file compared to the length the file is referred to as the real time factor (RTF).

Unless agreed otherwise with Speechmatics, the following behaviour will be considered acceptable use of the Cloud Services ASR (for V2 customers only). Speechmatics reserve the right to change the rate limits at any time in order to ensure continuity of service for all customers of the Cloud.

a. The Customer shall limit the rate of submission of files to a maximum of 2 jobs per second with a maximum of 100 jobs in progress at any one time.
b. The Customer shall limit the rate of polling for the status of submitted jobs to a maximum of 20 queries per second (across all jobs).

Ready to talk to Speechmatics?

Speak to Sales