Phonetic-based keyword search engine for audio streams, enabling search in massive libraries in seconds
Octopus Speech-to-Index is a search engine that indexes speech directly without converting it into text, enabling a keyword search within audio and video files.
Octopus Speech-to-Index finds any keyword, including proper names or slang, without knowing the exact spelling, removing the limitations of automated transcription solutions.
o = pvoctopus.create(access_key)metadata =o.index_audio_file(path)matches = o.search(metadata,phrases)Build with Python
Octopus o = new Octopus.Builder().setAccessKey(accessKey).build(appContext);OctopusMetadata metadata =o.indexAudioFile(path);HashMap <String, OctopusMatch[]>matches = o.search(metadata,phrases);Build with Android
let o =Octopus(accessKey: accessKey)let metadata =o.indexAudioFile(path: path)let matches = o.search(metadata: metadata,phrases: phrases)Build with iOS
const octopus =await OctopusWorkerFactory.create(accessKey,octopusIndexCallback,octopusSearchCallback);octopus.postMessage({command: "index",input: audioSignal,});octopus.postMessage({command: "search",metadata: octopusMetadata,searchPhrase: searchText,});Build with JavaScript
pv_octopus_t *octopus = NULL;pv_octopus_init(access_key,model_path,&octopus);pv_octopus_index_file(octopus,audio_path,&indices,&num_indices_bytes);pv_octopus_search(octopus,indices,num_indices_bytes,phrase,&matches,&num_matches);Build with C
Enterprises use automated transcription to find keywords and phrases in the audio and video libraries, despite not being built for this purpose. Automated transcription struggles with homophones and cannot transcribe words if not in the dictionary.
Octopus Speech-to-Index uses an acoustic-based search, achieving much higher accuracy than a generic transcription engine.
Monetize your content, monitor conversations, or ensure compliance without the limitations of automatic transcription.
Reduce errors by four times compared to Google Speech-to-Text. The open-source benchmark shows Octopus Speech-to-Index is the right tool for the job and outperforms the workarounds.
Make audio and video files searchable in seconds. Compared to Mozilla DeepSpeech, Octopus Speech-to-Index processes voice data 50 times faster while returning ten times more accurate results.
Protect sensitive information, such as call center recordings with personal data or legal depositions with confidential information. Automated transcription APIs send voice data to a 3rd party cloud to process it, while Octopus Speech-to-Index processes voice data anywhere.
The best way to learn about Octopus Speech-to-Index is to use it!
Start NowSpeech-to-Index, also known as audio indexing, speech indexing, and acoustic indexing, is a technique that makes audio automatically searchable and discoverable. As it performs searches based on phonetics, it’s also known as phonetic search, phonetic-based search, and acoustic search. It allows quick searches and rapid access to audio content. Picovoice built Octopus Speech-to-Index as a response to market demand. It indexes even massive audio and media libraries as Google indexes websites and returns keyword search results.
Octopus Speech-to-Index is built for finding keywords and phrases, whereas speech-to-text is for generic transcription. Given the maturity of text indexing algorithms, transcribing voice to text and then performing a search based on text seem like a good workaround to many. However, speech-to-text has limitations in correctly identifying these proper nouns and homophones. (Katia Leighton vs. Katja Layton and fair vs. fare). Acknowledging speech-to-text limitations, Picovoice built an acoustic-based phonetic search engine, Octopus Speech-to-Index, dedicated to finding keywords and phrases in audio libraries with high accuracy and speed.
Octopus Speech-to-Index enables many use cases, media asset management, legal e-discovery, dialogue search, and social media listening .
Octopus Speech-to-Index offers multilingual support for English, French, German, Italian, Japanese, Korean, Portuguese, Spanish, and more.
Contact Picovoice Sales and tell us about the opportunity, including the use case, requirements, and project details.
Picovoice docs, blog, Medium posts , and GitHub are great resources to learn about voice AI, Picovoice technology, and how to enhance speech quality. Picovoice also offers GitHub community support to all Free Plan users.