Guidelines and quotas
Amazon Transcribe has several guidelines to follow in order to achieve optimal results. There are also quotas that may impact your transcriptions; some of these can be increased. Refer to the following sections for details.
Supported Regions
For a list of AWS Regions where Amazon Transcribe is available, see Amazon Transcribe Endpoints and Quotas in the Amazon Web Services General Reference.
Guidelines
For best results:
-
Use a lossless format, such as FLAC or WAV, with PCM 16-bit encoding.
-
Use a sample rate of 8,000 Hz for low-fidelity audio and 16,000 Hz (or higher) for high-fidelity audio.
Amazon Transcribe may temporarily store your content to continuously improve the quality of its analysis
models. See the Amazon Transcribe FAQ
If you don't need to process all of your transcription jobs concurrently, use Job queuing. This enables
Amazon Transcribe to keep track of your transcription jobs and process them when slots are
available. You can request an increase to the job queue bandwidth ratio to run more
transcription jobs. The quota for the transcription jobs in your job queue is the
product of the number of transcription jobs you can run concurrently and the bandwidth
ratio. For example, if you have a bandwidth ratio of 5 and a quota of
100 for the number of transcription jobs you can run concurrently then
you can have 500 transcription jobs in your job queue.
Quotas
You can request a quota increase for the following resources:
| Resource | Default |
|---|---|
| Number of concurrent batch transcription jobs | 250 |
| Number of concurrent batch transcription jobs (call analytics) | 100 |
| Job queue bandwidth ratio | 0.9 |
| Number of concurrent HTTP/2 streams for streaming transcription | 25 |
| Number of StartStreamTranscriptionWebSocket requests | 25 |
| Total number of vocabularies per account | 100 |
| Number of pending vocabularies | 10 |
| Number of concurrently training custom language models | 3 |
| Total number of custom language models per account | 10 |
| Number of categories per account (call analytics) | 200 |
| Number of rules per category (call analytics) | 20 |
The following operations limits can also be increased upon request:
| Operation | Transactions per second |
|---|---|
| StartTranscriptionJob | 25 |
| StartStreamTranscription | 25 |
| GetTranscriptionJob | 30 |
| DeleteTranscriptionJob | 5 |
| ListTranscriptionJobs | 5 |
| CreateVocabulary | 10 |
| UpdateVocabulary | 10 |
| DeleteVocabulary | 5 |
| GetVocabulary | 20 |
| ListVocabularies | 5 |
| StartCallAnalyticsJob | 10 |
| GetCallAnalyticsJob | 20 |
| ListCallAnalyticsJobs | 5 |
| DeleteCallAnalyticsJob | 5 |
| CreateCallAnalyticsCategory | 10 |
| UpdateCallAnalyticsCategory | 10 |
| DeleteCallAnalyticsCategory | 5 |
| GetCallAnalyticsCategory | 20 |
| ListCallAnalyticsCategories | 5 |
For information about requesting a quota increase, see AWS Service Quotas in the Amazon Web Services General Reference.
The following quotas cannot be changed:
| Description | Quota |
|---|---|
| Audio file length | 4:00:00 (four) hours (14,400 seconds) |
| Audio file size | 2 GB |
| Audio file size (call analytics) | 500 MB |
| Size of a custom vocabulary | 50 KB |
| Length of a custom vocabulary phrase | 256 characters |
| Size of a vocabulary filter | 50 KB |
| Number of vocabulary filters | 100 |
| Number of channels for channel identification | 2 |
| Number of days job records are retained | 90 |
| Minimum audio file duration | 500 milliseconds (ms) |
