Speech encoding using a pre-encoded database

An apparatus comprising:

• a processor,
• memory coupled with the processor;
• and
• instructions stored in the memory and executable by the processor to cause the apparatus to:
• receive a digital representation of an audio signal;
• identify, based at least in part on receiving the digital representation of the audio signal, a database that is pre-encoded offline according to a coding standard and that comprises a quantity of digital representations of other audio signals, wherein the quantity of digital representations of other audio signals satisfies a set of thresholds, the set of thresholds comprising one or more of a power level threshold, a sampling frequency threshold, or a bit depth threshold;
• encode offline the digital representation of the audio signal using a machine learning scheme and information from the database pre-encoded according to the coding standard to match one or more frames of the digital representation of the audio signal to one or more frames from the database, wherein the encoded digital representation of the audio signal comprises one or more pre-encoded frames from the database that correspond to the one or more frames that match the digital representation of the audio signal;
• generate a bitstream of the digital representation that is compatible with the coding standard and irrespective of online encoding based at least in part on encoding the digital representation of the audio signal offline;
• and
• output a representation of the bitstream.

An apparatus comprising:

• means for receiving a digital representation of an audio signal;
• means for identifying, based at least in part on receiving the digital representation of the audio signal, a database that is pre-encoded offline according to a coding standard and that comprises a quantity of digital representations of other audio signals, wherein the quantity of digital representations of other audio signals satisfies a set of thresholds, the set of thresholds comprising one or more of a power level threshold, a sampling frequency threshold, or a bit depth threshold;
• means for encoding offline the digital representation of the audio signal using a machine learning scheme and information from the database pre-encoded according to the coding standard to match one or more frames of the digital representation of the audio signal to one or more frames from the database, wherein the encoded digital representation of the audio signal comprises one or more pre-encoded frames from the database that correspond to the one or more frames that match the digital representation of the audio signal;
• means for generating a bitstream of the digital representation that is compatible with the coding standard and irrespective of online encoding based at least in part on encoding the digital representation of the audio signal offline;
• and
• means for outputting a representation of the bitstream.