Parameter summary

access_token

A required access token that you use to establish an authenticated connection with the WebSocket interface. For more information, see Open a connection.

acoustic_customization_id

An optional customization ID for a custom acoustic model that is adapted for the acoustic characteristics of your environment and speakers. By default, no custom model is used. For more information, see Using a custom acoustic model for speech recognition.

audio_metrics

An optional boolean that indicates whether the service returns metrics about the signal characteristics of the input audio. By default (false), the service returns no audio metrics. For more information, see Audio metrics.

background_audio_suppression

An optional float between 0.0 and 1.0 that indicates the level to which background audio and side conversations are to be suppressed in the input audio. The default is 0.0, which provides no suppression of background audio. For more information, see Background audio suppression.

base_model_version

An optional version of a base model. The parameter is intended primarily for use with custom models that are updated for a new base model, but it can be used without a custom model. The default value depends on whether the parameter is used with or without a custom model. For more information, see Making speech recognition requests with upgraded custom models.

character_insertion_bias

An optional float between -1.0 and 1.0 that indicates whether the service is biased to recognize shorter (negative values) or longer (positive values) strings of characters when developing transcription hypotheses. By default, the service uses a default bias of 0.0. The value that you specify represents a change from a model's default. For more information, see Character insertion bias.

Content-Type

An optional audio format (MIME type) that specifies the format of the audio data that you pass to the service. The service can automatically detect the format of most audio, so the parameter is optional for most formats. It is required for the audio/alaw, audio/basic, audio/l16, and audio/mulaw formats. For more information, see Specifying an audio format.

customization_weight

An optional double between 0.0 and 1.0 that indicates the relative weight that the service gives to words from a custom language model versus words from the base vocabulary. The default differs for different types of models. You can specify a value either when a custom model is trained or when it is used for speech recognition. For more information, see Using customization weight.

end_of_phrase_silence_time

An optional double between 0.0 and 120.0 that indicates the pause interval at which the service splits a transcript into multiple final results if it encounters silence. By default, the service uses a pause interval of 0.8 seconds for all languages other than Chinese, for which it uses an interval of 0.6 seconds. For more information, see End of phrase silence time.

grammar_name

An optional string that identifies a grammar that is to be used for speech recognition. The service recognizes only strings that are defined by the grammar. You must specify both the name of the grammar and the customization ID of the custom language model for which the grammar is defined. For more information, see Using a grammar for speech recognition.

inactivity_timeout

An optional integer that specifies the number of seconds for the service's inactivity timeout. Inactivity means that the service detects no speech in streaming audio. The default is 30 seconds. Use -1 to indicate infinity. For more information, see Inactivity timeout.

interim_results

An optional boolean that directs the service to return intermediate hypotheses that are likely to change before the final transcript. By default (false), interim results are not returned. Interim results are available only with the WebSocket interface. For more information, see Interim results.

keywords

An optional array of keyword strings that the service spots in the input audio. By default, keyword spotting is not performed. For more information, see Keyword spotting.

keywords_threshold

An optional double between 0.0 and 1.0 that indicates the minimum threshold for a positive keyword match. By default, keyword spotting is not performed. For more information, see Keyword spotting.

language_customization_id

An optional customization ID for a custom language model that includes terminology from your domain. By default, no custom model is used. For more information, see Using a custom language model for speech recognition.

low_latency

An optional boolean that indicates whether the service is to produce results more quickly at the possible expense of transcription accuracy. By default (false), low latency is not enabled. For more information, see Low latency.

max_alternatives

An optional integer that specifies the maximum number of alternative hypotheses that the service returns. By default, the service returns a single final hypothesis. For more information, see Maximum alternatives.

model

An optional model that specifies the language in which the audio is spoken and the rate at which it was sampled: broadband/multimedia or narrowband/telephony. By default, en-US_BroadbandModel is used. For more information, see Using a model for speech recognition.

processing_metrics

An optional boolean that indicates whether the service returns metrics about its processing of the input audio. By default (false), the service returns no processing metrics. For more information, see Processing metrics.

processing_metrics_interval

An optional float of at least 0.1 that indicates the interval at which the service is to return processing metrics. If the processing_metrics parameter is true, the service returns processing metrics every 1.0 seconds by default. For more information, see Processing metrics.

profanity_filter

An optional boolean that indicates whether the service censors profanity from a transcript. By default (true), profanity is filtered from the transcript. For more information, see Profanity filtering.

redaction

An optional boolean that indicates whether the service redacts numeric data with three or more consecutive digits from a transcript. By default (false), numeric data is not redacted. If you set the redaction parameter to true, the service automatically forces the smart_formatting parameter to be true, and it disables the keywords, keywords_threshold, max_alternatives, and (for the WebSocket interface) interim_results parameters. For more information, see Numeric redaction.

smart_formatting

An optional boolean that indicates whether the service converts dates, times, numbers, currency, and similar values into more conventional representations in the final transcript. For US English, the feature also converts certain keyword phrases into punctuation symbols. By default (false), smart formatting is not performed. For more information, see Smart formatting.

smart_formatting_version

An optional boolean that indicates whether the service converts dates, times, numbers, currency, and similar values into more conventional representations in the final transcript. For more information, see Smart formatting version.

speaker_labels

An optional boolean that indicates whether the service identifies which individuals spoke which words in a multi-participant exchange. If you set the speaker_labels parameter to true, the service automatically forces the timestamps parameter to be true. By default (false), speaker labels are not returned. For more information, see Speaker labels.

speech_detector_sensitivity

An optional float between 0.0 and 1.0 that indicates the sensitivity of speech recognition to non-speech events in the input audio. The default is 0.5, which provides a reasonable level of sensitivity to non-speech events. For more information, see Speech detector sensitivity.

split_transcript_at_phrase_end

An optional boolean that indicates whether the service splits a transcript into multiple final results based on semantic features of the input such as sentences. The service bases its understanding of semantic features on the base language model, which can be further influenced by custom language models and grammars. By default (false), the service produces no semantic splits. For more information, see Split transcript at phrase end.

timestamps

An optional boolean that indicates whether the service produces timestamps for the words of the transcript. By default (false), timestamps are not returned. For more information, see Word timestamps.

Transfer-Encoding

An optional value of chunked that causes the audio to be streamed to the service. By default, audio is sent all at once as a one-shot delivery. For more information, see Audio transmission.

word_alternatives_threshold

An optional double between 0.0 and 1.0 that specifies the threshold at which the service reports acoustically similar alternatives for words of the input audio. By default, word alternatives are not returned. For more information, see Word alternatives.

word_confidence

An optional boolean that indicates whether the service provides confidence measures for the words of the transcript. By default (false), word confidence measures are not returned. For more information, see Word confidence.

X-Watson-Learning-Opt-Out

IBM Cloud

An optional boolean that indicates whether you opt out of the default request logging that IBM performs to improve the service for future users. To prevent IBM from accessing your data for general service improvements, specify true for the parameter. If you opt out, the service logs no user data from your requests, saving no audio or text to disk. You can also opt out at the account level. For more information, see Request logging.

X-Watson-Metadata

An optional string that associates a customer ID with data that is passed for recognition requests. The parameter accepts the argument customer_id={id}. By default, no customer ID is associated with the data. For more information, see Information security.