Google Inc.

Douglas Eck

Jay Yagnik

Hrishikesh Balkrishna Aradhye

David A Ross

Thomas Chadwick Walters

Christopher Tim Althoff

Ron J Weiss

Ryan Michael Rifkin

Samy Bengio

Adam L Berenzweig

Jay Michael Ponte

Jason Edward Weston

Tomas Izo

Huazhong Ning

Sally A Goldman

Rich Gossweiler

Richard Gossweiler

Mohamad Tarifi

Geremy A Heitz, Iii

Richard Francis Lyon

Karen Simonyan

Amarnag Subramanya

Mohammad Norouzi

Fernando Pereira

Sander Etienne Lea Dieleman

Thang Minh Luong

Peter Junteng Liu

Jesse Engel

Adam Roberts

Colin Abraham Raffel

Jennifer Ann Gillenwater

Garth Griffin

Cinjon Resnick

Generating a playlist may include designating a seed track in an audio library; identifying audio tracks in the audio library having constructs that are within a range of a corresponding construct of the seed track, where the constructs for the audio tracks are derived from frequency representations of the audio tracks, and the corresponding construct for the seed track is derived from a frequency representation of the seed track; and generating the playlist using at least some of the audio tracks that were identified.

Generating a playlist

A processor determines metadata associated with an audio track. The processor identifies categories that are related to the audio track based on the metadata. The processor determines rankings for the categories that are related to the audio track. The ranking is indicative of a relevance of a particular category to the audio track. The processor performs a query to identify visual media for one or more of ranked categories. The visual media is related to the audio track. The processor generates a visual presentation for the audio track by selecting at least some of the visual media to include in the visual presentation.

Creating a video for an audio file

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating a target sequence including a respective output at each of multiple output time steps from respective encoded representations of inputs in an input sequence. The method includes, for each output time step, starting from the position, in the input order, of the encoded representation that was selected as a preceding context vector at a preceding output time step, traversing the encoded representations until an encoded representation is selected as a current context vector at the output time step. A decoder neural network processes the current context vector and a preceding output at the preceding output time step to generate a respective output score for each possible output and to update the hidden state of the decoder recurrent neural network. An output is selected for the output time step using the output scores.

Sequence processing using online attention

The user feedback module receives from a user device a selection of a seed media item. The seed media item includes a seed label. The user feedback module further generates a list, the list includes the seed label and a set of related labels based on the seed label. The seed label and each related label include a separate label weight. The user feedback module also identifies multiple media items associated with an associate label. The associate label includes at least one of the seed label or a related label of the first set of related labels. The user feedback module also transmits to the user device a first media item from the multiple media items. The user feedback module adjusts the label weight of the associate label based on a user feedback on the first media item.

Adaptive music and video recommendations

Systems and methods described herein relate to automation of video creation for an associated audio file or musical composition. In particular, a video can be generated for the audio file that includes images and videos that are compelling and contextually relevant to, and technically compatible with, the audio file.

The present disclosure provides systems and methods that include or otherwise leverage a machine-learned neural synthesizer model. Unlike a traditional synthesizer which generates audio from hand-designed components like oscillators and wavetables, the neural synthesizer model can use deep neural networks to generate sounds at the level of individual samples. Learning directly from data, the neural synthesizer model can provide intuitive control over timbre and dynamics and enable exploration of new sounds that would be difficult or impossible to produce with a hand-tuned synthesizer. As one example, the neural synthesizer model can be a neural synthesis autoencoder that includes an encoder model that learns embeddings descriptive of musical characteristics and an autoregressive decoder model that is conditioned on the embedding to autoregressively generate musical waveforms that have the musical characteristics one audio sample at a time.

Generating music with deep neural networks

A user feedback module, executable by a processing device from memory is disclosed. The user feedback module receives from a user device a selection of a seed media item. The seed media item includes a seed label. The user feedback module further generates a list, the list includes the seed label and a set of related labels based on the seed label. The seed label and each related label include a separate label weight. The user feedback module also identifies multiple media items associated with an associate label. The associate label includes at least one of the seed label or a related label of the first set of related labels. The user feedback module also transmits to the user device a first media item from the multiple media items. The user feedback module adjusts the label weight of the associate label based on a user feedback on the first media item.

Techniques for generating a playlist include: receiving acoustic information, obtaining seed information based on the acoustic information, identifying audio tracks in an audio library based on the seed information, and generating the playlist using at least some of the identified audio tracks.

Generating a playlist based on input acoustic information

A system and methods for automatically detecting temporal music trends by observing music consumption by users of online services, for example, social networks, and user sharing habits. In some embodiments, the system and methods gather music consumption patterns (e.g., downloading, listening, sharing or the like) of users, including music identifiers for a track, album, or playlist in a user's music library and time stamps that indicate consumption times corresponding to the music identifiers. A temporal trends detection engine determines music of interest to users by analyzing music consumption patterns of users, user interests and tastes in music, and social affinity between users. A recommendations engine automatically generates and transmits recommendations of music determined by the temporal trends detection engine to be of interest to users.

System and methods for detecting temporal music trends from online services

A system identifies a set of initial segments of a time-based data item, such as audio. The segments can be defined at regular time intervals within the time-based data item. The initial segments are short segments. The system computes a short-timescale vectorial representation for each initial segment and compares the short-timescale vectorial representation for each initial segment with other short-timescale vectorial representations of the segments in a time duration within the time-based data item (e.g., audio) immediately preceding or immediately following the initial segment. The system generates a representation of long-timescale information for the time-based data item based on a comparison of the short-timescale vectorial representations of the initial segments and the short-timescale vectorial representations of immediate segments. The representation of long-timescale information identifies an underlying repetition structure of the time-based data item, such as rhythm or phrasing in an audio item.

Correlation-based method for representing long-timescale structure in time-series data

A system and method for providing an application programming interface for interfacing with audio files within a social networking service are provided. The method includes receiving, at the one or more computing devices, an ordered list of social networking contacts associated with a member of the social networking service. The ordered list of social networking contacts is arranged according to a ranking of the social networking contacts. The method also includes receiving a list of audio files associated with the member of the social networking service. The method also includes receiving a set of instructions. The set of instructions is configured to be implemented with the ordered list of social networking contacts and the list of audio files. The method also includes interfacing with the ordered list of social networking contacts or the list of audio files according to the set of instructions to modify the list of audio files.

Application programming interface for audio recommendation, discovery, and presentation within a social network

Techniques are disclosed for producing a collaborative recording of an audio event. An online server or service identifies participating mobile devices with recording capabilities that are available for recording at least a portion of the audio event. The online server or service determines the locations of the potential participating mobile devices, and identifies ranges of frequencies to be recorded by each of the participating mobile devices. The individual recordings are then compiled into a final collaborative recording.

Collaborative audio recording of an event by multiple mobile devices

A system and method for recommending audio files to member of social network are provided. Method includes receiving representation of first set of audio files and representation of set of social networking contacts associated with member, and representation of one or more second sets of audio files. Each second set of audio files is associated with at least one social networking contact in set of social networking contacts associated with member. Method also includes determining first set of characteristics based on representation of first set of audio files. Method also includes determining second set of characteristics based on representation of one or more second sets of audio files. Method also includes generating recommendation for new audio file based on first set of characteristics and second set of characteristics. New audio file is not in first set of audio files. Method also includes providing indication of recommendation for new audio file.

Recommending a new audio file to a member of a social network

Sound representations and winner-take-all codes of auditory spectra are used in the identification of audio content. A transformation component converts a set of sound frames from audio content into a set of spectral slices. A spectral encoder component encodes the spectral slices of auditory spectra into winner-take-all codes with a winner-take-all hash function. An identification component identifies which spectral dimension of a subset of spectral dimensions within a spectral slice has highest spectral value according to the winner-take-all codes. Reference audio content is determined to be similar or matching to the audio content based on the winner-take-all codes.

Growing community of inventors

Palo Alto, CA, United States of America

Douglas Eck