backchannel-classifier 0.4.0

backchannel classifier - detect backchannels vs real responses in thai and japanese asr output

backchannel classifier

detects backchannel responses vs real user input for voice ai systems. supports thai and japanese (aizuchi).

install

pip install backchannel-classifier

usage

from backchannel_classifier import is_backchannel

# thai (default)
is_backchannel("ครับ")                    # (True, 0.91)
is_backchannel("ไม่ครับ")                 # (False, 0.01)
is_backchannel("ใช่ แต่ว่า")              # (False, 0.01)

# japanese
is_backchannel("はい", lang="ja")         # (True, 0.99)
is_backchannel("そうですね", lang="ja")    # (True, 0.99)
is_backchannel("予約したいです", lang="ja") # (False, 0.0001)

# direct import
from backchannel_classifier.jp import is_backchannel_ja
is_backchannel_ja("なるほど")              # (True, 0.99)

returns (is_backchannel: bool, confidence: float).

why

voice bots using asr → llm → tts pipelines need to distinguish between backchannels (acknowledgment sounds that should be ignored) and real responses that need processing. simple exact matching fails on asr variants and misses edge cases.

approach

gradient boosting classifier with handcrafted language-specific features. key idea: strip known backchannel components from the text, measure what's left (remaining_ratio). if nothing remains, it's a backchannel.

thai (26 features)

feature	importance
remaining_ratio	0.9098
has_request	0.0406
has_negation	0.0274
particle_ratio	0.0108

• polite particle detection (ครับ/ค่ะ/จ้ะ variants)
• backchannel sound patterns (อืม/อ๋อ/เออ with tone variants)
• question/negation/request/continuation markers
• handles asr misspellings (ค่า→ค่ะ, คับ→ครับ, อื้ม→อืม)

japanese (27 features)

feature	importance
remaining_ratio	0.7765
remaining_len	0.0484
katakana	0.0347
word_count	0.0325
kanji_ratio	0.0206

• core aizuchi (はい/ええ/うん/そう)
• agreement, understanding, surprise, filler, reaction markers
• question/continuation/request/negation/verb negative indicators
• handles asr elongation variants (はーーい, えーーー)

results

thai

• 99.49% f1 (5-fold cv)
• test suite: 94/94 (100%)

japanese

• 98.37% f1 (5-fold cv)
• test suite: 119/119 (100%)

test coverage

thai (94 cases)

backchannels (49): ครับ, ค่ะ, อืม, ใช่, อ๋อ, เหรอ, ฮัลโหล, asr variants... real responses (45): สวัสดีครับ, ไม่ครับ, ราคาเท่าไหร่ครับ, edge cases (ใช่ แต่ว่า, ครับ แล้วก็)...

japanese (119 cases)

aizuchi (63): はい, うん, そうですね, なるほど, へー, まじで, えーと, すごい, 承知しました, compounds... real responses (56): ありがとうございます, いくらですか, 予約したいです, edge cases (はい、質問があります, そうですね、でも...)...

testing

python3 -m pytest tests/ -v

files

• backchannel_classifier/__init__.py - thai classifier + unified api
• backchannel_classifier/jp.py - japanese classifier
• train.py - thai training script
• train_ja.py - japanese training script
• tests/test_classifier.py - thai test suite (94 cases)
• tests/test_classifier_ja.py - japanese test suite (119 cases)

requirements

• python 3.8+
• scikit-learn
• numpy

memory

~3.7 MB per language model, lazy-loaded. if you only use thai, japanese model is never loaded (zero overhead).