Categories of Vocalizer parameters

Installation parameters

These parameters are seldom changed after installation. However, system administrators can override the default values. These variables have the following precedence:

Vocalizer determines a default value during initialization.
System administrators can override defaults with an environment variable of the same name.
The application can override the environment variable with the parameter in a Vocalizer configuration file (tts_config.xml).

Parameter	Description	Default
NUANCE_DATA_DIR	Directory for storing log, cache, and utterance files.	Windows: C:\ProgramData\Nuance\Enterprise Linux: /var/local/Nuance
PID	Process ID for the process running Vocalizer.	(defined by the operating system)
TMPDIR	Sets a storage location for temporary files.	(defined by the operating system)
USER	Current user’s login name.	(defined by the operating system)
VOCALIZER_SDK	Installation directory for Vocalizer for Enterprise.	Windows:C:\Program Files (x86)\Nuance\Vocalizer for Enterprise Linux:/usr/local/Nuance/Vocalizer_for_Enterprise
VOCALIZER_USERCFG	Specifies a Vocalizer configuration file (tts_config.xml) to be used for Vocalizer for Enterprise. The nserver.nvs.VocalizerUserCfgFile command-line option overrides this variable.	(user-defined)
VOCALIZER_VOICE_PATH	Path to voices. Set when specifying a non-default location for Vocalizer for Enterprise voices.	Windows:C:\Program Files (x86)\Nuance\Vocalizer for Enterprise Linux:/usr/local/Nuance/Vocalizer_for_Enterprise

Internet fetch parameters

These parameters control Internet fetch behaviors.

Parameter	Description	Default
inet_accept_cookies	Whether to accept HTTP cookies.	1 (accepted)
inet_cache_control_max_age	Internet fetch HTTP/1.1 max-age directive for controlling caching.	-1
inet_cache_control_max_stale	Internet fetch HTTP/1.1 max-stale directive for controlling caching.	-1 (disabled)
inet_extension_rules	Rules for mapping file name extensions to MIME content types.	(multi-value default)
inet_timeout_download	Default timeout for downloading a URI (from open through the final byte).	30000 (milliseconds)
inet_timeout_io	Default timeout for reading or writing a block of data over a web server connection.	30000 (milliseconds)
inet_timeout_open	Default timeout for opening a connection to a web server.	30000 (milliseconds)
inet_user_agent	User agent name in HTTP/1.1 headers.	Vocalizer for Enterprise/6.2

You can use these recognizer parameters to set up a proxy.

Parameter	Description	Default
inet_proxy_server_port	Port of the proxy server to use.	8080
inet_proxy_server	IP address of an proxy server to use.	(empty value; no proxy used)

Internet cache parameters

These parameters tune the cache.

Parameter	Description	Default
cache_directory	Directory name for the disk cache.	$NUANCE_DATA_DIR/system/cache/nvn
cache_entry_exp_time	Expiration time for unused disk cache entries.	3600 (seconds)
cache_entry_max_size	Maximum size of a single disk cache entry.	20 (MB)
cache_low_watermark	Maximum size of the cache after removing expired entries.	400 (MB)
cache_timeout_open	Timeout for opening a cache entry.	5000 (milliseconds)
cache_total_size	Maximum size of the disk cache.	500 (MB)
cache_unlock_entries_at_startup	Reserved for future use.	true

Internet peer authentication parameters

These parameters control peer authentication when using TLS.

Parameter	Description	Default
inet_ssl_ca_certificates_file	A file containing one or more sequential PEM-encoded public CA certificates.	(none)
inet_ssl_certificate_file	Specifies the PEM-encoded certificate.	(none)
inet_ssl_private_key_file	Specifies the PEM-encoded private key.	(none)
inet_ssl_verify	Enables peer authentication.	0
inet_ssl_verify_depth	Limits the depth of the certificate chain for validation.	2 (accommodates one intermediate CA)

Automatic language detection parameters

Use these parameters to control how Vocalizer automatically switches the synthesis language for blocks of text where the language is unknown.

Parameter	Description	Default
language_identifier_scope	Enables automatic language identification and sets its scope.	user-defined
language_identifier_languages	Permissible languages for automatic language identification in precedence order.	(all installed voices that support language identification)
language_identifier_mode	Mode for automatic language identifiication during the session.	rejection

Licensing parameters

Vocalizer requires licenses from a License Manager.

Parameter	Description	Default
cpr_license_dynamic_ports	Maximum number of CPR-only licenses that can be dynamically acquired.	0
cpr_license_ports_overdraft_thresh	Threshold to generate warnings when approaching the CPR license limit.	0 (no warnings)
cpr_license_ports	Number of licenses (CPR-only) for concatenated prompt recordings to acquire at startup.	0
dynamic_license_expiration_time	Sets the expiration timer for relinquishing unused dynamically acquired licenses.	300 (seconds)
license_explicit	Indicates whether licenses are reserved per SPEAK request, or per TTS engine instance.	false (per instance)
license_servers	TCP/IP addresses of License Manager hosts for Vocalizer.	27000@localhost
tts_license_dynamic_ports	Maximum number of full TTS licenses that can be dynamically acquired after exhausting all full TTS licenses acquired at startup (defined by tts_license_ports).	0
tts_license_ports_overdraft_thresh	Threshold at which warnings will be logged to indicate that the system is near the full TTS license limit.	0 (no warnings)
tts_license_ports	Number of full TTS licenses to acquire at startup.	4 (licenses)

Logging parameters

Parameter	Description	Default
application_name	The application name written to the call log.	(empty string)
event_log_file_base_name	Sets the base name for call logs.	${TMPDIR}/nvn-event-${USER}-
event_log_file_enabled	Enables call logging.	0 (disabled)
event_log_file_max_size	Sets the maximum size of the call log.	50 (MB)
event_log_input_text_file_base_name	Input text logging file base name.	${TMPDIR}/nvn-event-${USER}-
event_log_input_text_file_max_size	Sets the maximum size of the input text log file.	50 (MB)
event_log_input_text_max_capture	When event logging is enabled, the maximum number of speak requests where the input text is captured and logged at once.	-1 (unlimited)
log_cb_enabled	Whether to log errors via the callback mechanism.	1 (enabled)
log_file_base_name	Error and diagnostic log file base name.	${TMPDIR}/nvn-${USER}-
log_file_enabled	Whether to log errors and diagnostics to a Vocalizer-written log file.	0 (disabled)
log_file_max_size	Log file maximum size.	50 (MB)
log_level	Diagnostic log level for Vocalizer messages.	0 (normal errors and warnings)
log_to_system_log	Whether to log errors to the system log.	0 (disabled)
product_version	Returns the current product version string, such as "5.5.0".	(current product version)

Data security parameters

Parameter	Description	Default
secure_context	Sets security levels to protect confidential data from being written to log files or recorded in audio waveforms.	open
inet_ssl_enable_fips	Requires connections to secure servers to be FIPS-compliant.	disabled

Speak parameters

These parameters control the sound of synthesized speech.

Parameter	Description	Default
early_emission_allowed	Reduces latency when using xpremium-high voices.	false
escape_sequence	Alternative ESC character (ASCII 0x1B) for Vocalizer control sequences within the input text. (ESC continues to work even when this is set.)	\\!
marker_mode	Specifies the types of markers to deliver.	TTS_MRK_BOOK
rate	Speaking rate, where lower values are slower speaking rates.	50
ssml_validation	Sets the mode to evaluate SSML markup in input text.	strict
volume	Specifies the amplitude of TTS audio, where lower values are lower volume levels.	80 for Vocalizer API 100 for Microsoft SAPI API

Sentence detection parameters

Use these parameters to specify whether and how the ends of distinct sentences in input text are detected and cause pauses in synthesized speech.

Parameter	Description	Default
disable_eos_detection	Specifies whether to disable automatic end-of-sentence detection.	false
disable_final_sentence_silence	Specifies whether to suppress the final silence that follows input text.	false
wait_factor	Determines the duration of the pause inserted to mark the end of a sentence.	2

Tuning data parameters

Use these parameters for tuning speech output. They load tuning artifacts.

Parameter	Description	Default
activeprompt_dbs	ActivePrompt databases for tuning speech output.	(none)
dictionaries	User dictionaries to load for tuning speech output.	(none)
rulesets	Rulesets to load.	(none)

Voice selection parameters

These parameters describe the qualities of the voice selection.

Parameter	Description	Default
frequency	Voice frequency (sampling rate) as an enumerated type.	(not applicable)
language	Language for synthesis.	(none)
voice	Voice name for synthesis.	(none)
voice_age	Voice age.	Adult
voice_gender	Voice gender.	(empty string, all genders allowed)
voice_model	Force Vocalizer to use an older TTS technology when synthesizing the selected voice model.	(newest technology available)
voice_rescan_interval	Limit on how frequently Vocalizer rescans voice installation directories.	600 (seconds)

Categories of Vocalizer parameters

Related topics