Dynamic audio processing parameters with automatic speech recognition
A communication system includes a front-end audio gateway or bridge and a hands-free device. An automatic speech recognition platform accessible to the hands-free device provides or makes available one or more preprocessing schemes and/or acoustic models to the front-end audio gateway or bridge. The preprocessing schemes or acoustic models can be identified by or provided before a connection is established between the front-end audio gateway and the automatic speech recognition platform, when a connection occurs between the front-end audio gateway and the automatic speech recognition platform, and/or during a speech recognition session.
1. A communication system comprising:
an audio-gateway that converts input audio signals into a compatible form used by a receiving network;
a noise reduction module resident to the audio gateway configured to reduce in-vehicle echo and noise; and
a speech recognition engine remote from the audio gateway generates and transmits commands through a wireless network that cause the audio gateway to modify the audio gateway's noise reduction processing state in response to a recognized request at the audio gateway for an automated speech recognition;
where the noise reduction module applies knowledge of a reverberation time, knowledge of a plurality of microphone placement, and knowledge of driver's location stored in a data store to reduce the in-vehicle echo and noise.
2. A communication process comprising:
transferring a plurality of preprocessing schemes or acoustic models that can be implemented by an audio gateway through a short-range network;
comparing a spoken utterances to a grammar-based vocabulary to generate a recognition result and a confidence score at the audio gateway; and
selecting one of the plurality of preprocessing schemes or acoustic models by a automatic speech recognition process remote from the audio gateway based on the recognition result and the confidence score when the recognition result and the confidence score indicates a request for an automated speech recognition.
3. The communication process of claim 2 where the transfer of the plurality of preprocessing schemes or acoustic models occurs during a speech recognition session.
4. A communication system comprising:
an audio-gateway that converts input audio signals into a compatible form used by a receiving network;
a noise reduction module resident to the audio gateway configured to reduce in-vehicle echo and noise; and
a speech recognition engine remote from the audio gateway generates and transmits commands through a wireless network that cause the audio gateway to modify the audio gateway's noise reduction processing state in response to a request for an automated speech recognition
where the noise reduction module applies knowledge of a reverberation time, knowledge of a plurality of microphone placement, and knowledge of driver's location stored in a data store to reduce the in-vehicle echo and noise.
5. A communication process comprising:
transferring a plurality of noise or echo preprocessing schemes that are implemented by an audio gateway in response to commands received through a short-range network and the detection of a speech event at the audio gateway;
comparing a spoken utterances to a grammar-based vocabulary to generate a recognition result and a confidence score at the audio gateway; and
selecting one of the plurality of the nose or echo preprocessing schemes in response to an automatic speech recognition process remote from the audio gateway based on the recognition result and the confidence score when the recognition result and the confidence score indicates a request for automated speech recognition.