How do Credits Work?
What production credits are and how do they work?
At AudioStack the API usage is metered in production credits. Credits are based on the complexity of each task so some endpoints use more credits than others.
Below is a full list of services and their respective consumption in credits:
| As route | Request Type | Credits |
|---|---|---|
| Voice Intelligence Layer (per 10 seconds) | 0.5 | |
| Mastering Builder (per 10 seconds) | 0.5 | |
| /content/script | POST | 0.1 |
| /content/script | GET | 0 |
| /content/script | PUT | 0 |
| /content/script | DELETE | 0 |
| /content/script/{id} | GET | 0 |
| /content/script/{id} | DELETE | 0 |
| /content/script/{id}/{version} | GET | 0 |
| /content/script/{id}/{version} | DELETE | 0 |
| /content/scripts | GET | 0 |
| /content/scripts | DELETE | 0 |
| /content/list_projects | GET | 0 |
| /content/list_modules | GET | 0 |
| /content/file/create-upload-url | POST | 3 |
| /content/files | GET | 0 |
| /content/file/transfer-file | PUT | 1 |
| /content/file/id/{mediaId} | GET | 0 |
| /content/file/id/{mediaId} | DELETE | 0.25 |
| /speech/diction | GET | 0 |
| /speech/diction/custom | GET | 0 |
| /speech/diction/custom/item | PUT | 5 |
| /speech/diction/custom/item | DELETE | 0.25 |
| /speech/diction/custom/items | GET | 0 |
| /speech/predict | POST | 0.25 |
| /speech/predict/voices | GET | 0.25 |
| /speech/tts | POST | 0 |
| /speech/tts | GET | 0 |
| /speech/tts/{speechId} | DELETE | 0 |
| /speech/tts/{speechId} | GET | 0 |
| /speech/tts/preview | POST | 0.5 |
| /speech/tts/sync | POST | 5 |
| /speech/tts/reduce | POST | 1 |
| /speech/tts/annotate | POST | 1 |
| /speech/voice | GET | 0.25 |
| /speech/voice | POST | 0.25 |
| /speech/voice/{name} | GET | 0.25 |
| /speech/voice/select | POST | 5 |
| /voice/parameter | GET | 0.25 |
| /production/sound/template | POST | 25 |
| /production/sound/template | GET | 3 |
| /production/sound/template | PUT | 5 |
| /production/sound/template/{name} | DELETE | 0.25 |
| /production/sound/parameter | GET | 0.5 |
| /production/sound/segment | POST | 5 |
| /production/mix | POST | 5 |
| /production/mix/{productionId} | GET | 0.5 |
| /production/mix/{productionId} | DELETE | 0.25 |
| /production/mixes | GET | 0.5 |
| /production/mix/presets | GET | 1 |
| /delivery/encoder | POST | 5 |
| /delivery/encoder/presets | GET | 0.5 |
| /delivery/connectors/julep | POST | 5 |
| /voice/parameter | GET | 0.25 |
| Instant Voice Cloning | 300 |
Voice Providers
Here is an overview of the usage per voice provider per 1 minute.
| Text-to-speech voice provider (ID in API) | Credits per 1 minute of speech |
|---|---|
| Aflorithmic Messner (msnr) | 1.5 |
| Azure by Microsoft (azure) | 1 |
| CereProc (cerevoice) | 1.2 |
| Eleven Labs (elevenlabs) | 9 |
| Google (google) | 1 |
| IBM (ibm) | 1.2 |
| Polly by Amazon (polly) | 1 |
| Resemble (resemble) | 9 |
| WellSaid Labs (wellsaid) | 9 |
| OpenAI (openai) | 1.5 |
| PlayHT (playht) | 1.5 |
| Narakeet (narakeet) | 1.5 |
| Speechify (speechify) | 9 |
| Respeecher (respeecher) | 1.5 |
| Cartesia (cartesia) | 1.5 |
Which voice providers are the best for my use case?
Here is an overview of all TTS voice providers currently available on AudioStack
Updated about 2 months ago