This document will walk you through how to use the RESTful API for the Speechmatics SaaS. This document will show you
This section shows:
The Speechmatics SaaS supports the following languages. Your ability to use any or all of the languages will depend on what languages you are contracted to use.
Speechmatics takes a global first approach to our languages. In a single language pack we aim to support many different accents and dialects. This simplifies your workflow when selecting which language to use, not requiring you to know which accent is being spoken in your audio up-front. With this approach we still achieve very high accuracy compared to accent specific language packs.
Please note any languages outside this list are not explicitly supported. Only one language can be processed within each request. Each language above has a two-letter ISO639-1 code that must be provided for any transcription request.
The Speechmatics SaaS also supports specialized language packs that enhance the requested transcription language with optimization for a particular field. This is particularly useful for increasing the accuracy for domains that have specific terminology. The domain packs build on our global languages to give the best accuracy.
|Finance||(en)||Improve accuracy for audio containing financial terms such as those found in earnings calls or financial broadcast|
Refer to documentation here for more details on how to set the domain parameter.
The Speechmatics SaaS supports the following file types for transcription:
The list above is exhaustive - any file format outside the list above is explicitly not supported.
The supported size limit for jobs is 2 hours of audio or 1 GB file size. Any larger or longer files may be rejected.
Audio files, transcripts, and configuration data are stored in the Speechmatics SaaS for 7 days. Any request to retrieve a transcript or file more than 7 days after it was processed will receive a HTTP 404 error message and a status of
You can delete audio or transcripts in advance of this 7 day period - how to do so is documented in the How-To Guide
This section talks through the Pre-Requisites to authenticate to the Speechmatics SaaS
Speechmatics Support will provide you a unique Authorization Token. This must be used with any interaction with the Speechmatics SaaS to authenticate to the service. Any interaction without this token will receive a
HTTP 401 - Unauthorized response.
The Authorization Token must be passed in the header of any request to the Speechmatics SaaS. The Token itself conforms to OAuth2 protocols.
It is your responsibility to store the Authorization Token securely and ensure it is not mishandled. It is recommended that you store and provide access to the Token on the principle of least privilege. If you believe that your Token has been compromised, please contact Speechmatics Support.
If you are using the trial Speechmatics endpoint, please note your authorization token may be time limited. If you try and use the token after the trial period has expired, you will receive a
HTTP 401 - Unauthorized response.
Below are the supported endpoints.You should use the hostname for the region you are contracted to use:
If you want to use a different region please contact email@example.com. If you try to use a hostname for a region you are not contracted to use, that job will be unsuccessful
If you want to receive notifications from the Speechmatics SaaS, you will need to whitelist the following egress IP addresses. You should only whitelist the IP addresses from the geo region you are contracted to use
The list is below:
Speechmatics SaaS applies rate limiting and fair queueing to provide a consistently high quality of service to all users.
If you make a large number of requests in a short period of time, some of these requests may fail with the response
HTTP 429 - Rate Limited. To minimise the possibility of encountering rate limiting errors, we recommend that you do not exceed the following rates:
Aside from rate limiting, there is no limit to the number of jobs that you can submit. However, Speechmatics SaaS applies a fair queueing policy which means that if you have a large number of jobs in progress at one time, the most recently submitted jobs may take longer to complete.
If you require a service which exceeds the recommendations above, please contact firstname.lastname@example.org
Please note: all timestamps from the Speechmatics SaaS are in ISO 8601 format, and use Coordinated Universal Time (UTC).
The API uses a standardised authentication method, with a wide range of supported output formats and has a structured JSON configuration that is easier to extend in the future.
For the full API specification please refer to the API Reference section.