By voice alone, Italian Torontonians with a high degree of ethnic orientation (EO) were correctly identified 78% of the time in Nagy et al.’s (2020) ethnolinguistic perception study. Comparing production and perception, we discern indexical features of the Toronto Italian English ethnolinguistic repertoire (TIER) on which these accurate judgments rest. Our analysis considers Italian substrate features (Krämer 2009), Canadian English features (Hoffman and Walker 2010), and features noted in the ethnolectal literature (Szakay 2012, Newman and Wu 2011). We examine these features’ distribution in the sociolinguistic interviews of 8 second-generation high-EO Italians, 8 low-EO Italians, and 8 British background speakers. Augmenting these findings with commentary from Italian Torontonian group interviews, we identify linguistic markers of Toronto Italian identity. A mixed-effects model predicting Euclidean Distance for 20,000+ /ow/ and /ey/ tokens indicates that high-EO Italians produce the vowels significantly more monophthongally than low-EO Italians and British-background speakers. A mixed effects model predicting spectral tilt measures for 70,000+ vowel tokens shows the same division. No significant inter-group distinctions for Canadian English variables emerged. Supporting the reallocation of minority language features to new social functions (Gnevsheva 2020), monophthongal /ow/ and /ey/ may have initially transferred from the comparable vowels in Italian but now index Italian identity for second generation Torontonians, regardless of the speaker’s Italian fluency. Identified as indexical of other ethnic groups (Szakay 2012, Newman and Wu 2011), modal phonation may be a less established cue for social distinction (Bucholtz and Hall 2004). Monophthongization was salient to participants in our group interviews; modal voice was not.