Overview
Ensure key phrases are picked up by the language recognition model and correct faulty transcriptions.
Overview
PolyAI uses advanced Automatic Speech Recognition (ASR) models to convert spoken language into written text. To improve the recognition of critical domain-specific terms, Jupiter introduces the Keyphrase Boosting feature.
Key benefits
- Improved SLU accuracy: Increases the likelihood of accurately recognizing critical terms.
- Domain adaptability: Ensures the agent recognizes industry-specific terminology.
- Enhanced user experience: Reduces misunderstandings and errors, leading to more effective interactions.
- Easy management: Manage and update keyphrases in the UI.
How it works
Keyphrase Boosting biases the ASR model toward recognizing specific words and phrases. By curating a list of keyphrases relevant to your domain, you can improve transcription accuracy for those terms, resulting in better inputs for the Language Learning Model (LLM) and improved agent performance.
Getting started
Configuring Keyphrase Boosting
- Access the Speech Recognition page: Navigate to the Speech Recognition section.
- Manage keyphrases:
- In the Keyphrase Boosting tab, add, edit, or remove keyphrases.
- Use the Keyphrase column to input domain-specific terms.
- Set bias strength:
- Adjust the bias strength for each keyphrase using the slider.
- Options range from Default (moderate priority) to Maximum (high priority).
- Save changes: Once configured, the updated keyphrases will be applied immediately to improve ASR recognition.
Bias strength configuration
- Default: Balances recognition accuracy with overall ASR performance.
- Maximum: Prioritizes keyphrases for improved accuracy but may impact general performance.
Important: When both global and local biasing are applied, local settings take precedence.
Overview
PolyAI uses advanced Automatic Speech Recognition (ASR) models to convert spoken language into written text. To improve the recognition of critical domain-specific terms, Jupiter introduces the Keyphrase Boosting feature.
Key benefits
- Improved SLU accuracy: Increases the likelihood of accurately recognizing critical terms.
- Domain adaptability: Ensures the agent recognizes industry-specific terminology.
- Enhanced user experience: Reduces misunderstandings and errors, leading to more effective interactions.
- Easy management: Manage and update keyphrases in the UI.
How it works
Keyphrase Boosting biases the ASR model toward recognizing specific words and phrases. By curating a list of keyphrases relevant to your domain, you can improve transcription accuracy for those terms, resulting in better inputs for the Language Learning Model (LLM) and improved agent performance.
Getting started
Configuring Keyphrase Boosting
- Access the Speech Recognition page: Navigate to the Speech Recognition section.
- Manage keyphrases:
- In the Keyphrase Boosting tab, add, edit, or remove keyphrases.
- Use the Keyphrase column to input domain-specific terms.
- Set bias strength:
- Adjust the bias strength for each keyphrase using the slider.
- Options range from Default (moderate priority) to Maximum (high priority).
- Save changes: Once configured, the updated keyphrases will be applied immediately to improve ASR recognition.
Bias strength configuration
- Default: Balances recognition accuracy with overall ASR performance.
- Maximum: Prioritizes keyphrases for improved accuracy but may impact general performance.
Important: When both global and local biasing are applied, local settings take precedence.
Overview
Accurate transcription is vital for effective agent responses, particularly in specialized industries like healthcare, legal, and technical support. The Transcript Corrections feature enables custom string corrections to ensure domain-specific terms and misheard phrases are accurately transcribed, enhancing the quality of inputs for the LLM.
Key benefits
- Improved accuracy: Correct common ASR misinterpretations.
- Domain-specific customization: Tailor corrections for unique terminology.
- Streamlined workflow: Configure corrections directly in the PolyAI UI.
- Enhanced user experience: Ensure precise transcription for reliable interactions.
- Efficient LLM inputs: Deliver cleaner text for improved response quality.
How it works
Transcript Corrections use string matching and regex patterns to manage misinterpretations. When a match is detected:
- The system replaces the incorrect word or phrase with the specified correction.
- Corrections can be applied to the entire transcript or specific portions.
Getting started
Configuring Transcript Corrections
- Access the Speech Recognition page: Navigate to the Speech Recognition section.
- Open the Transcript Corrections tab: From here, you can create, edit, or delete correction rules.
- Create a new correction:
- Click Add Correction to open the creation form.
- Define the correction:
- Regex: Specify the regular expression to identify the misinterpreted phrase.
- Replacement: Enter the correct term or phrase.
- Replacement type:
- Full transcript: Replaces the entire transcript if matched exactly.
- Partial transcript: Replaces only the matching portion.
- Save changes: Apply your corrections immediately.
Example configuration
| Attribute | Value | |
|---|---|---|
| ID | stop_inappropriate | |
| Regex | `/\b(offensiveWord1 | offensiveWord2)\b/i` |
| Replacement | [Correct term or phrase] | |
| Replacement type | Partial transcript |
This setup ensures phrases like “offensiveWord1” are replaced, maintaining accuracy and professionalism.