Speech codecs
AssigneeNokia Corporation
InventorJari Makinen
Life of this patent
Estimating the term…
Abstract
A method and apparatus include a voice activity detection module configured to detect silent frames, and a codec mode selection module configured to determine a codec mode. The voice activity detection module includes a receiver configured to receive a frame, a first determiner configured to determine a first set of parameters from the frame, and a providing unit configured to provide the first set of parameters to the codec mode selection module. The codec mode selection module includes a second determiner configured to determine a second set of parameters in dependence on the first set of parameters, and a selector configured to select a codec mode in dependence on the second set of parameters.
Claims14
A method, comprising:
- receiving a frame at a voice activity detection module;
- determining, at the voice activity detection module, a first set of parameters from the frame;
- providing the first set of parameters to a codec mode selection module;
- determining, at the codec mode selection module, a second set of parameters in dependence on the first set of parameters;
- and
- selecting a codec mode to encode the frame at the codec mode selection module in dependence on the second set of parameters to determine a coding mode,
- wherein the first set of parameters comprises at least one of a sub-band energy level and a background noise estimate, and
- wherein the second set of parameters comprises at least one of normalized energy levels in each sub-band of the frame, an average frequency of the frame, and an energy level difference between the current and previous frames.
An apparatus, comprising:
- a voice activity detection module configured to detect silent frames;
- and
- a codec mode selection module configured to determine a codec mode, wherein the voice activity detection module comprises
- a receiver configured to receive a frame,
- a first determiner configured to determine a first set of parameters from the frame, wherein the first set of parameters comprises at least one of a sub-band energy level and a background noise estimate, and
- a provider configured to provide the first set of parameters to the codec mode selection module,
- wherein the codec mode selection module comprises
- a second determiner configured to determine a second set of parameters in dependence on the first set of parameters, wherein the second set of parameters comprises at least one of normalized energy levels in each sub-band of the frame, an average frequency of the frame, and energy level difference between the current and previous frames, and
- a selector configured to select a codec mode in dependence on the second set of parameters.
An apparatus, comprising:
- voice activity detection means for detecting silent frames;
- and
- codec mode selection means for determining a codec mode,
- wherein the voice activity detection means comprises
- receiving means for receiving a frame,
- first determining means for determining a first set of parameters from the frame, wherein the first set of parameters comprises at least one of a sub-band energy level and a background noise estimate, and
- providing means for providing the first set of parameters to the codec mode selection means,
- wherein the codec mode selection means comprises
- second determining means for determining a second set of parameters in dependence on the first set of parameters, wherein the second set of parameters comprises at least one of normalized energy levels in each sub-band of the frame, an average frequency of the frame, and energy level difference between the current and previous frames, and
- selecting means for selecting a codec mode in dependence on the second set of parameters.
Dependencies are inferred from the claim text.