Differences
This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
draft [2009/07/15 15:52] ptacek |
draft [2009/07/16 09:49] ptacek |
||
---|---|---|---|
Line 4: | Line 4: | ||
+ | ====== Description of Czech Companion November Demonstrator ====== | ||
- | ====== Description of Czech Companion | + | The Czech version of the Companion |
- | The Czech version of Companion deals with the Reminiscing about User's Photos scenario taking advantage of data recorded in first phase of the project. The basic architecture | + | The dialog |
- | however | + | |
- | + | ||
- | photopal domena, nahranej korpus, ze na to sou dafy (reusing SHEFF DM intergrated through Inamode Relayer | + | |
- | typy odpovedi a zpusob jejich implementace, | + | |
- | NLP server s tectomt, ASR/TTS/SR client, connected over network | + | |
- | XXX JPta | + | |
+ | Our DAFs covering selected topics contain not only Companion replies mined from the corpora, but also new human-authored assessments, | ||
+ | {{user: | ||
Line 27: | Line 24: | ||
- | ===== Speech Reconstruction (WP 5.1 ???) ===== | + | |
+ | ===== Speech Reconstruction (WP 5.2) ===== | ||
features: omit filler phrases, remove irrelevant speech events, handle false starts, repetitions, | features: omit filler phrases, remove irrelevant speech events, handle false starts, repetitions, | ||
performance indicator: BLEU score between actual output and manually reconstructed sentences from corpora (T5.2.1), baseline: Moses with default settings | performance indicator: BLEU score between actual output and manually reconstructed sentences from corpora (T5.2.1), baseline: Moses with default settings | ||
+ | |||
+ | |||
Line 37: | Line 37: | ||
===== Morphology Analyzer and POS tagging (WP 5.2) ===== | ===== Morphology Analyzer and POS tagging (WP 5.2) ===== | ||
- | features: coverage of photo-pal domain, domain adapted tagger | + | features: coverage of photo-pal domain, domain adapted tagger |
performance indicator: OOV rate, accuracy | performance indicator: OOV rate, accuracy | ||
+ | |||
Line 45: | Line 46: | ||
features: induce dependencies and labels | features: induce dependencies and labels | ||
performance indicator: accuracy (correctly induced edges, labels) | performance indicator: accuracy (correctly induced edges, labels) | ||
- | v tipu je natrenovat MacDonnalda na dialog datech, ten task je do M42, ted ne. | ||
Line 52: | Line 52: | ||
===== Semantic Parsing (WP 5.2) ===== | ===== Semantic Parsing (WP 5.2) ===== | ||
- | features: assignment of semantic roles (69 roles), coordinations, | + | features: assignment of semantic roles (69 roles), coordinations, |
performance indicator: accuracy (correctly induced edges, labels) | performance indicator: accuracy (correctly induced edges, labels) | ||
Line 63: | Line 63: | ||
===== Named Entities Recognition (WP 5.2) ===== | ===== Named Entities Recognition (WP 5.2) ===== | ||
- | features: detect person names, geographical locations, | + | features: detect person names, geographical locations, |
performance indicator: f-measure | performance indicator: f-measure | ||
Line 70: | Line 70: | ||
===== Dialog Act Tagging (WP 5.2) ===== | ===== Dialog Act Tagging (WP 5.2) ===== | ||
- | features: domain tailored | + | features: domain tailored |
performance indicator: accuracy | performance indicator: accuracy | ||
+ | |||
+ | |||
+ | |||
Line 79: | Line 82: | ||
===== Dialog Manager (WP 5.3) ===== | ===== Dialog Manager (WP 5.3) ===== | ||
- | features: | + | features: |
- | Manually created | + | manual creation of DAFs covering following topics: |
performance indicator: acceptability - manual evaluation of actions selected by DM | performance indicator: acceptability - manual evaluation of actions selected by DM | ||
- | ===== Natural Language Generation (WP 5.4) ===== | ||
- | features: morphological adjustments, | ||
- | performance indicator: BLEU score | ||
- | ====== AZ PO LISTOPADU ====== | ||
- | ===== Syntactic Parsing | + | ===== Natural Language Generation |
- | features: | + | features: |
- | performance indicator: | + | performance indicator: |
- | ===== Sentiment Analysis (WP 5.2) ===== | ||
- | features: za tohle bych vydaval klasifikator, | ||
- | performance indicator: f-measure | ||
- | ===== Complete System Evaluation | + | ===== Emotional TTS (WP 5.5) ===== |
- | T5.2.7 tohle zminuje, nick webb to pro nas asi neudela | + | features: emotions will be expressed implicitly, through the usage of communicative functions; new female voice database was recorded for this purposes |
- | performance indicator: | + | performance indicator: |
- | ===== advances ===== | ||
- | advances in Czech NLU (on reconstructed spoken data): 300-500vet(?) rucne anotovat pos, a-tree, t-tree, IE predicates, Named Entities, DA pro eval in-domain testy after Nov. | + | ===== Emotional Avatar Integration |
- | pos ? analyzovat, generovat a kontrolovat ' | + | features: new Czech female voice with affective features will be integrated with the TID avatar |
+ | performance indicator: subjective evaluation of the naturalness and the ability to convey emotions | ||
+ |