Speech Services
teliSpeech University
A set of 8 training modules are currently available in our catalogue. They cover pre-sales, core technology, tools and integration aspects of our know-how.
They help you to:
- Develop your speech recognition know-how
- Improve the quality of your speech applications
- Increase your autonomy regarding our tools
- Explore new possibilities
Training sessions can be carried out on-site or inter-company, they can also be personalized to fit your needs.
Pre-sales Training
To provide pre-sales engineers with an overview of projects using speech recognition and ensure that they are able to identify the necessary subjects involved in the pre-sales phase.
Duration:
1 day including a workshop session
Audience:
- Project managers
- Pre-sales engineers
- Client account managers
Pre requisites:
- Project life cycle and quality control
- Drafting of technical offers
- Voice recognition
Topics:
- An overview of teliSpeech's functionalities and state of the art voice recognition
- The project life cycle and adoption of metric of follow-up quality for your offer
- The main points that need to be covered to finalize your offer, in particular
- The collection of information on what already exists
- Identification of specific constraints or limitations in the existing business/infrastructure
- Definition of the full functional requirements
- Sizing of services related to the use of ASR technology
Please note: this training does not cover other aspects of systems using voice recognition. Thus other subjects covered in OnMobile's training catalogue are not included in this training.
Delivery method:
presentations and practical sessions
Languages:
English, French
Speech Technology Standards Training
To provide a complete overview of speech standards: MRCP, VoiceXML, SRGS, SISR.
Duration:
1 day
Audience:
- Voice applications developers
- Engineers responsible for the integration
- Project managers
Pre requisites:
XML, TCP/IP
Topics:
- Overview of a speech platform
- Introduction to VoiceXML
- Speech recognition in VoiceXML
- SRGS, a W3C standard for speech recognition grammars
- MRCP, the IETF standard for integration speech resource providers
Delivery method:
Presentations and practical sessions
Languages:
English, French
Integration Workshop Training
To provide a good understanding of speech recognition concepts and teliSpeech architecture before starting an integration of teliSpeech. Understand the differences between a native C++ integration and a MRCP integration
Duration:
1 day
Audience:
- Engineers responsible for the integration
- Product development managers
Pre-requisites
none
Topics:
- Speech recognition concepts
- teliSpeech architecture
- Presentation of the MRCP standard
- Presentation of the teliSpeech C++ API
- Open topics relative to your specific needs
Delivery methods:
presentations
Languages:
English, French
Administration of teliSpeech Training
To provide those implicated in the integration phase of teliSpeech with a detailed knowledge of its installation, maintenance deployment and the follow up of the applications which have been deployed on the voice platform.
Duration:
1 day
Audience:
- Technology integrators
- IT managers
- System architects
- System adminstrators
- Customer project managers
- Application developers
Pre-requisites:
- Exploitation system follow up and technical quality control (optional)
- SNMP process supervision (strongly recommended)
- Program installation on Windows and Linux (mandatory)
- Basic knowledge of the TPC/IP and UDP protocols (strongly recommended)
- Logging function
Topics:
- General operation of teliSpeech
- teliSpeech architecture
- Full installation of teliSpeech
- teliSpeech configuration
- teliSpeech deployment and the links between different teliSpeech processes
- The main principles of maintenance and redundancy for teliSpeech
- Levels of logging within teliSpeech and use of log information
- teliSpeech monitoring with SNMP traps
- How to use the teliSpeech documentation and support
Please note: this training does not cover other aspects of systems using voice recognition. So, other subjects covered in OnMobile's training catalogue are not included in this training program.
Delivery methods:
Presentations and practical sessions
Languages:
English, French
Grammar Design and Optimisation Training
To provide a general level of knowledge and gain pratical competencies in grammar development for the teliSpeech engine, including all phases of specification up until deployment.
Duration:
2 days
Audience:
- Voice application developers
- Engineers responsible for the integration
- Project managers
Pre-requisites:
- Project life cycle and quality control of voice applications
- Speech technology in particular speech recognition
- XML/VoiceXML and Javascript programming
Topics:
- The design steps and grammar compilation
- The efficient development of grammars in the view to quality and performance control
- The tuning of teliSpeech settings, in particular the ASR server
- The range and the use of the teliSpeech toolkit
- The improvement of voice recognition within the general tuning process
Please note: this training does not cover other aspects of systems using voice recognition. Thus ergonomics, specification and VoiceXML programming are not included in this training program
Delivery methods:
Presentations and practical sessions
Languages:
English, French
Statistical Language Model Training
To provide a general level of knowledge and gain pratical competencies in building statistical language model used in continuous speech recognition.
Duration:
2 days
Audience:
- Voice applications developers
- Project managers
Pre-requisites:
- Grammar development and optimisation training
- Project life cycle and quality control
- Speech technology, in particular speech recognition
- Script language programming
Topics:
- Theory of statistical language model (SLM)
- SLM design process
- SLM grammar compilation
- The range and use of the teliSpeech SLM toolkit
- The improvement of SLM recognition performance within the general tuning process
Delivery methods:
Presentations and practical sessions
Languages:
English, French
Automatic Call Routing Training
To provide a general level of knowledge and gain practical competencies in building an automatic classification model for call routing application.
Duration:
1 day
Audience:
- Voice application developers
- Project managers
Pre-requisites
- statistical language models training is a mandatory pre-requisites for this training
Topics:
- Advantages of building an automatic call routing application
- Principles of automatic classification
- Process for building an automatic classification model
- Testing of the classification model
- Evaluating overall performance of an natural language call routing system
- Updating a natural language call routing application
- Optimizing a natural language call routing application
Delivery methods:
Presentations and practical sessions
Languages:
English, French
Acoustic Tuning Training
To provide a general level of knowledge and gain pratical competencies in building acoustic models which will increase speech recognition performance.
Duration:
2 days
Audience:
- Voice application developers
- Project managers
Pre-requisites:
- Grammar development and optimisation training
- Project life cycle and quality control
- Speech technology, in particular speech recognition
- Script language programming
Topics:
- Principles of acoustic tuning
- Acoustic tuning process
- Building speech database for tuning acoustic models
- The range and use of the teliSpeech tuning toolkit
- Evaluating performance improvement with the created acoustic models
Delivery methods:
Presentations and practical sessions
Languages:
English, French