Method for interoperation between adaptive multi-rate wideband (AMR-WB) and multi-mode variable bit-rate wideband (VMR-WB) codecs

A method for encoding a speech signal according to a first speech coding scheme so that it can be decoded according to a second speech coding scheme, the speech signal comprising active speech periods during which there is active speech and inactive speech periods during which there is no active speech, the first speech coding scheme having a first set of available coding modes, each of said first set of coding modes having an associated encoding bit-rate, the second speech coding scheme having a second set of available coding modes including a discontinuous transmission coding mode in which silence descriptor frames are generated during inactive speech periods, the method comprising:

• receiving an input speech signal for encoding according to the first speech coding scheme;
• applying a speech frame derived from the input speech signal to a voice activity detection function to determine whether the speech frame is an active speech frame containing active speech or an inactive speech frame that does not contain active speech;
• when it is determined that the input speech frame is an inactive speech frame, performing a determination operation according to a predetermined rule to specify whether, according to the second speech coding scheme, the inactive speech frame is to be encoded as a silence descriptor frame;
• and
• when it is determined that the input speech frame is to be encoded as a silence descriptor frame, encoding the input speech frame using a first predetermined encoding mode selected from said first set of available encoding modes that has an encoding bit-rate sufficiently high to allow encoding of the input speech frame with a number of bits compatible with a silence descriptor frame according to the second speech coding scheme;
• when it is determined that the input speech frame is not to be encoded as a silence descriptor frame, encoding the input speech frame using a second predetermined encoding mode selected from said first set of encoding modes.

An apparatus for encoding a speech signal according to a first speech coding scheme so that it can be decoded according to a second speech coding scheme, the speech signal comprising active speech periods during which there is active speech and inactive speech periods during which there is no active speech, the first speech coding scheme having a first set of available coding modes, each of said first set of coding modes having an associated encoding bit-rate, the second speech coding scheme having a second set of available coding modes including a discontinuous transmission coding mode in which silence descriptor frames are generated during inactive speech periods, the apparatus comprising:

• an input for receiving a speech signal for encoding according to the first speech coding scheme;
• a voice activity detector for determining whether a speech frame derived from said speech signal can be classified as an active speech frame containing active speech or an inactive speech frame that does not contain active speech;
• an inactive speech frame processing unit operable to perform a determination operation on a speech frame classified as inactive according to a predetermined rule to specify whether, according to the second speech coding scheme, the inactive speech frame is to be encoded as a silence descriptor frame;
• and
• an encoding unit responsive to the determination operation performed by said inactive frame processing unit, operable to encode the input speech frame using a first predetermined encoding mode selected from said first set of available encoding modes when it is determined that the input speech frame is to be encoded as a silence descriptor frame, said first predetermined encoding mode having an encoding bit-rate sufficiently high to allow encoding of the input speech frame with a number of bits compatible with a silence descriptor frame according to the second speech coding scheme and operable to encode the input speech frame using a second predetermined encoding mode selected from said first set of encoding modes when it is determined that the input speech frame is not to be encoded as a silence descriptor frame.

A circuit comprising:

• an input for receiving a speech signal for encoding according to a first speech coding scheme for decoding according to a second speech coding scheme, the speech signal comprising active speech periods during which there is active speech and inactive speech periods during which there is no active speech, the first speech coding scheme having a first set of available coding modes, each of said first set of coding modes having an associated encoding bit-rate, the second speech coding scheme having a second set of available coding modes including a discontinuous transmission coding mode in which silence descriptor frames are generated during inactive speech periods;
• a voice activity detector for determining whether a speech frame derived from said speech signal can be classified as an active speech frame containing active speech or an inactive speech frame that does not contain active speech;
• an inactive speech frame processing unit operable to perform a determination operation on a speech frame classified as inactive according to a predetermined rule to specify whether, according to the second speech coding scheme, the inactive speech frame is to be encoded as a silence descriptor frame;
• and
• an encoding unit responsive to the determination operation performed by said inactive frame processing unit, operable to encode the input speech frame using a first predetermined encoding mode selected from said first set of available encoding modes when it is determined that the input speech frame is to be encoded as a silence descriptor frame, said first predetermined encoding mode having an encoding bit-rate sufficiently high to allow encoding of the input speech frame with a number of bits compatible with a silence descriptor frame according to the second speech coding scheme and operable to encode the input speech frame using a second predetermined encoding mode selected from said first set of encoding modes when it is determined that the input speech frame is not to be encoded as a silence descriptor frame.

An apparatus comprising:

• means for inputting a speech signal to encode the speech signal according to a first speech coding scheme for decoding according to a second speech coding scheme, wherein the speech signal comprises active speech periods during which there is active speech and inactive speech periods during which there is no active speech, the first speech coding scheme having a first set of available coding modes, each of said first set of coding modes having an associated encoding bit-rate, the second speech coding scheme having a second set of available coding modes including a discontinuous transmission coding mode in which silence descriptor frames are generated during inactive speech periods;
• means for detecting voice activity in a speech frame derived from the input speech signal to determine whether the speech frame is an active speech frame containing active speech or an inactive speech frame that does not contain active speech;
• means to perform a determination operation according to a predetermined rule when it is determined that the input speech is an inactive speech frame, to determine if, according to the second speech coding scheme, the inactive speech frame is to be encoded as a silence descriptor frame;
• and
• means for encoding the input speech frame, when it is determined that the input speech frame is to be encoded as a silence descriptor frame, using a first predetermined encoding mode selected from said first set of available encoding modes that has an encoding bit-rate sufficiently high to allow encoding of the input speech frame with a number of bits compatible with a silence descriptor frame according to the second speech coding scheme;
• and
• means for encoding the input speech frame, when it is determined that the input speech frame is not to be encoded as a silence descriptor frame, using a second predetermined encoding mode selected from said first set of encoding modes.